Professional Documents
Culture Documents
INTRODUCTION
Archaeologists have allowed the present-day border between the United States and
Mexico to influence the study of the Northwest-Southwest1 region. Researchers on the U.S. side
of the frontier have let this recent line in the landscape guide the way they have constructed the
pre-Hispanic world. Archaeologists have characterized Northern Mexico, for the most, as a vast
strip of mostly barren landscape between the U.S. Southwest and Mesoamerica. Early research
tended to interpret the area in terms of the well-known archaeological cultures to the north and
south. Recently, investigators in Northern Mexico have begun the overwhelming task of
providing an accurate culture-history that represents the true individuality of this area. The
research recently undertaken at the site of Cerro de Trincheras (Figure 1), in Sonora, Mexico, of
The site of Cerro de Trincheras overlooks the Magdalena River Valley by the modern day
town of Trincheras, Sonora. Researchers agree that this terraced hill is the largest and most
impressive “cerro de trincheras”2 (O’Donovan 1997, McGuire et al. 1994). This site is
1
I refer to the area typically referred to as the Southwest as the Northwest-Southwest to
emphasize the inclusion of areas south of the border. See Phillips 1989 for summary of the area.
2
The site type cerro de trincheras refers to large isolated terraced hills in the Northwest-
Southwest. They are found in Sonora, Chihuahua, and Arizona. The site type cross cuts the
Hohokam (Stacy 1974; Downum et al. 1994; Kirk 1994) , Casas Grandes (Di Peso 1974; Kirk
1994, Roney and Hard 1998), Trincheras (McGuire et al.1993), and Rio Sonora (Pailes 1974;
Doolittle 1984, 1988) cultural traditions.
1
Figure 1. The Site of Cerro de Trincheras
2
associated with the Trincheras tradition in northern Sonora and southern Arizona (Kirk 1994).
The site’s impressive character results from the more than 900 terraces constructed on it and the
variety of structures built on them. Cerro de Trincheras exhibits strong evidence of many
activities associated with village life. The radiocarbon dates from the site suggest that Cerro de
Trincheras was inhabited from A.D. 1300 to 1450 (McGuire and Villalpando 1998).
Dr. Randall McGuire and Elisa Villalpando directed archaeological excavations at Cerro
de Trincheras between January and May in 1995 and 1996. Villalpando and McGuire realized
the project as a cooperative effort between the State University of New York at Binghamton and
the , Centro Regional Sonora del Instituto Nacional de Antropología e Historia. The National
Science Foundation funded the project (NSF Grant SBR930224). The Cerro de Trincheras
Research Project was the first large scale excavation of a site associated with the Trincheras
tradition. Due to the site’s massive nature, only 1.5% of the site has been excavated to date. The
The construction of a basic culture-history for the Trincheras Tradition is at its initial
allow researchers to make comparisons on the basis of material regularities. Yet, a baseline
classification of Trincheras architecture has never been attempted. Until recently, not enough
data was available to create an architectural typology. The Cerro de Trincheras Research Project
provides for the first time the type of data needed to characterize Trincheras features. This thesis
of Trincheras culture.
My four goals in this thesis are: 1) to present a typology for the features found at Cerro de
3
Trincheras; 2) to describe the features that correspond to each type and show their organization
on the site; 3) to apply ethnographic and ethnohistorical data to reconstruct what the structures
might have looked like; 4) to compare these feature types generally with those in the Hohokam
region to the north. The remainder of this chapter is dedicated to contextualizing the main
questions and introducing the organization of this thesis. I use data gathered by the two
excavation seasons and subsequent analysis for this endeavor. Analysis permitted the
identification of 13 general feature types: circular stone structures, quadrangular stone structures,
Trincheras research has traditionally been undertaken beneath the shadow of Hohokam
investigation. In order to move away from this and consider the Trincheras tradition as a unique
cultural entity, it is helpful to compare and contrast the material record of these two groups. The
comparison allows archaeologists to appreciate the differences and similarities between these two
groups.
The culture historical theoretical approach guided early Hohokam researchers. The
culture history of the Hohokam prompted investigators to draw boundaries in the landscape to
demarcate the limits of this culture area. Archaeologists defined boundaries based on material
differences. The evident changes in the material record of the Hohokam sites to the south
prompted some archaeologists to interpret sites in southern Arizona and northern Sonora as
apparent retardation because of the distance from the Hohokam core in the Phoenix Basin.
4
Investigators interpreted the Trincheras tradition as an expression of this “backward” Hohokam.
However, at that point little research had been done in northern Mexico.
Emil Haury and Charles C. Di Peso developed theories to address the material change in the
as the Desert branch of the Hohokam (Haury 1976:8; 1950: 547). Di Peso explained the
differences in the material record as expressions of a different ethnicity, the O’otam (Martin and
Plog 1973:46). Both researchers sent students down to the Trincheras area to fit it into their
interpretations.
Since then, archaeologists have undertaken very little work in northern Mexico. As a
result, today we are conducting basic baseline research in the Trincheras area. This thesis
Chapter 2 summarizes the background of the site of Cerro de Trincheras. I divide the
chapter into three sections: environmental setting, previous research, and the culture history of
the Trincheras tradition. The previous research section pays special attention to the work that
In Chapter 3, I describe the Cerro de Trincheras research project. First, the chapter
includes a description of the site. In the site description section I include a general description of
the site and the areas excavated by the Cerro de Trincheras Research Project. Second, I
summarize the history of the project. The history of the project section includes dates,
organization and logistics of the Cerro de Trincheras Research Project. Finally, I outline the
5
methodology used for the excavation, analysis, and classification of features.
The features at Cerro de Trincheras are the subject of Chapter 4. I describe each of the 14
feature types found at Cerro de Trincheras. The definitions are supplemented by a detailed
summary of the data recovered during excavation and subsequent analysis. The chapter provides
the features. This chapter addresses two issues: the internal organization of feature types and
excavated. The superstructure section uses ethnohistoric and ethnographic sources to posit
possible forms for the perishable superstructures above circular stone structures, quadrangular
Chapter 6 uses the typology and results outlined in Chapter 4 to compare the architectural
features of Cerro de Trincheras with those of the Hohokam. The results show a difference
between a Hohokam village and Cerro de Trincheras despite the environmental similarities. The
results further emphasize the need for the study of the Trincheras tradition as a result of
historically situated and contingent actions of the pre-Hispanic inhabitants of the Mexican
Sonoran desert.
The final chapter presents a summary and conclusions of the study of features excavated
at the site of Cerro de Trincheras. Results of the comparison beg for further research that
interprets the Trincheras tradition for itself and not as an extension of the Hohokam.
6
CHAPTER 2: BACKGROUND INFORMATION
Many visitors and researchers have been captivated by Cerro de Trincheras. Although it
has captured the attention of many, very little is known about the people who built and lived on
this breathtaking site. Our recent excavations have helped to answer fundamental questions
about the Trincheras tradition. This chapter provides background information about the site.
First, I describe the environmental setting of Cerro de Trincheras. Next, I provide a summary of
relevant previous research at the site, the region and on other site cerros de trincheras sites.
Finally, I review the culture history for the Trincheras tradition of Sonora.
Natural Setting
environment requires people to deal with high temperatures and the sun’s radiation. Solutions to
living in a desert environment are often expressed arquitectonically. Thus, the description of the
natural setting focuses on factors that would influence feature types. The following summary of
the environmental setting is largely taken from Mc Guire & Villalpando 1995: 6-8.
Cerro de Trincheras (Figure 2) is an isolated volcanic hill that over looks the Magdalena
River valley in Northern Sonora. The Magdalena River Valley is found in the Sonoran desert
subdivision of the North American Desert. The region is in the Basin and Range physiographic
province. Researchers characterize this area by long mountain ranges that rise from areas of flat
desert.
7
Figure 2. Aerial Photo of Cerro de Trincheras
8
The temperature in the region often exceeds 100/F. Populations living in desert
environments must factor in the wind, heat and sun. Solutions to these environmental
considerations are often expressed architecturally. Builders in hot, dry climates construct
structures that deal with temperature, humidity, rate of air movement, and radiation from walls
floors ceilings and other surrounding surfaces (Saini 1973). These factors usually result in two
kinds of desert dwellings: structures with thick walls to absorb the suns radiation or dwellings
that have light permeable side walls that promote the circulation of air through the structure
A biannual precipitation pattern characterizes the Sonoran desert, resulting in both winter
and summer rain. Storm fronts from the Pacific cause the winter rains, while summer rains come
as monsoons. The monsoon rainfall pattern results in heavy precipitation in short bursts. Only
plants that do not require much water survive in this area. This limits the amount of trees
The region does provide a great deal of plant resources that could have been used to
construct the superstructure of dwellings at Cerro de Trincheras. The flora of the Magdalena
River Valley conforms to the Lower Colorado Valley subdivision of the Sonoran desert
(O’Donovan 1997: 82). Creosote and white bursage compose most of the vegetation found in the
plains. A variety of cacti grow in the area. Among them are cholla, saguaro, cardon, ocotillo,
and organ pipe cacti. These species of cacti are mostly found in the upper bajada areas. Saguaro
ribs and ocotillo branches are especially good materials for building sidewalls and roofs. The
upper bajadas are also the home to several species of small trees including mesquite, ironwood
and paloverde. In the alluvial river valleys, higher concentrations of mesquite and ironwood are
9
found. The site is ideally situated between the floodplain of the Magdalena River and the bajada
that descends from the Sierra Santa Rosa (O’Donovan 1997: 84). The pre-Hispanic inhabitants
of Cerro de Trincheras had access to the resources of the bajada vegetation and the riverine
features into it or manipulating it to create a level surface. The summer rains are ideal for the
creation of caliche. Caliche is a concentration of lime which solidifies desert soils. Caliche is
“created when calcium carbonate from surface deposits is carried in solution as water percolates
into soil, where evaporation results in its precipitation” (Stone 1986: 9).
The builders of Cerro de Trincheras took advantage of the building materials the hill itself
provided. The volcanic hill that they built the site on provides fine grained igneous rocks.
Andesite cobbles of varied sizes occur most commonly occur on the hill. Andesite cobbles are
Previous Research
The United States-Mexico international border should not have an impact on the study on
the study of the pre-Hispanic groups in Northwest Mexico and the U.S. Southwest. In the last
two decades, work on the northern side of the border multiplied because of contract archaeology.
Investigators can thus see a glaring contrast in what they know of the north of Mexico and what
they know of the U.S. Southwest. The effect the border has had is two fold: a lack of
archaeological projects in northern Mexico and bias in the way archaeologists have interpreted
pre-Hispanic groups that inhabited this large expanse. The lack of research in northern Mexico
has led to interpretations of northern Mexico as a peripheral area controlled by the “Southwest”,
10
to the north, or Mesoamerica, to the south.
One victim of this border phenomenon is the site of Cerro de Trincheras. The Cerro de
Trincheras research project is one of a handful of projects being conducted in Northern Mexico
which aim to remedy this situation. In this section I will outline the research done with respect
to Cerro de Trincheras to date. Additionally, I will present previous interpretations of the site to
Spanish explorer Captain Manje wrote the earliest reference to the site of Cerro de
Trincheras in 1694 (Burrus 1971: 178). He was the first to use the term “trincheras” for the
terraces. Though the term has been the subject of some confusion, the explorer used the term
“trincheras” to mean fortifications. Manje counted 100 terraces at the site and made his
The Jesuit Pfefferkorn (1883: 63) referred to cerros de trincheras in his description of
Sonora (Villalpando 1998: 3). Pfefferkorn recorded his experience living at the missions of Atil
and Cucurpe in the middle of the eighteenth century, describing cerros de trincheras sites as
vestiges of the wars between the Pimas, Opatas, Seris, and other nations. Pfefferkorn said these
groups used the hills as fortresses. The hills still have ruins of the trenches built as walls.
Pfefferkorn describes the walls as piles of rocks on top of each other beginning at the base of the
hill.
At the end of the nineteenth-century, explorers visiting Sonora reported the existence of
large hills with terraces and walls. First, Schumacher published his observations as an article
titled Ancient Fortifications in Sonora (Schumacher 1881). Next, Hamilton followed his lead by
publishing his descriptions of cerros de trincheras (Hamilton 1883). Neither Schumacher nor
11
Hamilton never actually visited the site.
The site of Cerro de Trincheras was first described extensively by McGee at the end of
the nineteenth-century. McGee was the first to record any of the features on the site. He
published three works which refer to the site (McGee, 1895, 1896, and 1898; Carmony and
Brown 1983). McGee (1896) described some circular stone structures, the spiral at the crest of
the hill (El Caracol, Spanish for the snail shell), and some large stone and bedrock engravings at
the site. McGee also thought the site must have served to defend against invaders from the
In 1910 Huntington visited cerro de trincheras sites in the Magdalena Valley. Huntington
cites cerros de trincheras as the proof of a climatic shift that caused a period of famine and
warfare (Huntington 1914: 73-74). He interpreted these sites as fulfilling military, agricultural
and religious functions (Huntington 1914: 67-70). Specifically, Huntington proposed that the
inhabitants of Cerro de Trincheras might have used the terraces to “cultivate special products,
such as grapes” that they watered using large water jars (Huntington 1914: 65-70). Huntington
first assigned the site of Cerro de Trincheras to the Southern Arizona culture he called the
1912). In this work Lumholtz counted twenty terraces at the site of Cerro de Trincheras. He
described circular stone structures and artifacts at the site. His extensive travels through the area
allowed Lumholtz to differentiate the function of Cerro de Trincheras from other cerro de
trincheras sites that he interpreted as fortifications. He saw the exposed southern side of the hill
as evidence that the site did not have a defensive purpose. The plethora of domestic materials
12
made the agricultural interpretation doubtful. This led Lumholtz to interpret the site as a
ceremonial center designed to serve some defensive needs (Lumholtz 1912: 142).
Monroe Amsden conducted a reconnaissance survey along the Rio Sonora in 1928.
During this survey, Amsden (1928: 47) described rectangular alignments of stones, two courses
high, measuring approximately 2.5 to 3 meters wide by 3 to 4 meters long. He interpreted the
cobble lines as foundations of pole and brush huts. Amsden noted that jacal structures were very
In 1929 Frank Midvale visited the Trincheras area and valley of Altar as part of a survey
of the Papaguería. Gladwin later used Midvale’s data to oppose the environmental deterministic
theories Huntington had proposed. They saw the cerro de trincheras as the “developmentally
retarded Hohokam” due to their distance from the Hohokam core in the Phoenix basin in Arizona
Informed by the Gladwinian interpretations of the Papaguería, Sauer and Brand (1931)
conduct an extensive survey of Northern Sonora. Their research led them to refute Huntington’s
theories on environmental determinism, because they noted that the fill of the terraces was not
agricultural. Sauer and Brand’s analysis of ceramic distributions led them to coin the term
Trincheras culture. Sauer and Brand defined the Trincheras culture as pre-Hispanic sites in the
Magdalena and Altar river valleys and some hills, containing mainly purple-on-red pottery,
characterized by terraces and “rock corrals”3 (Sauer and Brand 1931:107-109, 117-118). They
describe corrals as “heavy enclosures of rock, five or six feet high, circular or elliptical and of
3
I have defined “rock corrals” that many researchers describe at cerro de trincheras sites
as circular and quadrangular stone structures.
13
varying size” (Sauer and Brand 1931: 69). Sauer and Brand did not determine any house forms,
but describe stone foundations of as possible pithouses. They take issue with Amsden’s
interpretation stone alignments as possible jacales because these structures do not use rocks in
their construction. It appears that Sauer and Brand were describing stone structures while
Amsden was describing the stone alignments left by jacales. The heavy concentration of
Fletcher A. Carr (1935) completed his master thesis that is an overview of the “ancient
pueblo culture of northern Mexico.” His research brought Carr to Cerro de Trincheras. He
interpreted the site as a place of dwelling and “possibly protection”. He described the hill as
being covered with terraces that stand from two to ten feet high. Carr mentions the terraces vary
in length up to 300 yards. He goes on to say that very little soil covered the terraces. This leads
Carr to conclude that they did not serve an agricultural purpose. He based his habitational
interpretation on the presence of several rock rooms on the hill. He goes to describe one of these
rock rooms: “One room measured twelve and a half feet long by eight feet wide by five feet
Mexico. In this work he defined three cultural complexes in northern Chihuahua and Sonora,
Chihuahuan, Rio Chihuahuan and Trincheras and two pottery types for Trincheras tradition:
Brand later interpreted Trincheras as a separate cultural phenomenon from the Hohokam. He
attributed the material culture differences to an adversarial relationship between the Trincheras
and the Hohokam people (Brand 1935: 300; O’Donovan 1998: 38).
14
In the early 1930s Woodward (1936) visited the site of La Playa a few kilometers north of
Cerro de Trincheras. He identified a Trincheras shell industry which he differentiated from the
jewelry manufacturing center, that traded shell ornaments with the Hohokam. Woodward further
assigned the origins of some Trincheras and Hohokam stylistic elements to Mesoamerica.
Around this time, Gordon Ekholm (1939, 1940, 1947) conducted extensive surveys in
Sonora and northern Sinaloa. He recorded several cerros de trincheras in Sonora, though he
focused more on sites in Sinaloa. During his survey, Ekholm visited Cerro de Trincheras.
Ekholm’s stay at the site gave him the occasion to map El Caracol in plan view.
The late 1930s and early 1940s brought several researchers to the Papaguería of Arizona
and Sonora (Fraps 1936; Hoover 1941). All three researchers recorded several cerros de
trincheras sites in the Papagueria. Fraps recorded 114 circular rooms at the Black Stone Ruin in
Southern Arizona. Hoover’s work on the Papago reservation allowed him to create a list of
structural elements on cerro de trincheras. This list included the following architectural
elements: “terraces; small, circular walls; breastworks across approaches; and enclosed plazas”
The 50s brought new questions to the study of the Papaguería. Researchers studying the
Hohokam focused on the role of the Papaguería in the Hohokam framework, given the
differences in material culture. Emil Haury (1950: 547) interpreted the material record of areas
south of the Phoenix Basin, including the Trincheras culture as expressions of a desert adapted
branch of the Hohokam that he named the Desert Hohokam. Table 1 shows the traits Haury used
to differentiate the Desert branch from the River branch of the Hohokam. Haury saw the
15
material
Full troughed metate, well shaped Block m etate, some sh aping, not tro ughed, as a r ule
Great arra y of projec tile points, delica te workma nship Limited projectile point types, few in number,
workmanship poor
Few chop ping, scrap ing, and cutting to ols Abundance of roughly chipped chopping, scraping, and
cutting tools
Stone jew elry, abunda nt elaborate Stone and shell jewelry, rare and simple
Large-scale irrigation systems drawing water from Limited irrigation canals, designed to catch surface
streams runoff
Charles Di Peso had a different interpretation for the inhabitants of the Papaguería
partially influenced by sites east and south of the Tucson basin (McGuire 1988: 22). Di Peso saw
the differences as a consequence of ethnic conflict. Di Peso placed the O’otam (a branch of the
desert Cochise culture) at Snaketown and the Gila-Salt drainage until A.D. 1000. Then the
Hohokam, an intrusive band from Mexico, the Hohokam, took control over the region.
16
According to Di Peso, the Hohokam absorbed the O’otam culture. In the Papaguería the O’otam
survived as the Desert branch Hohokam (Martin and Plog 1973:146). Di Peso interpreted
Trincheras as derived from the O’otam. Specifically, Di Peso saw certain affinities between the
upper Pima (O’otam) and the Trincheras ceramic style and manufacture.
Haury and Di Peso wanted to incorporate Trincheras in their debates so each sent
researchers to Mexico. In the early 1960s, Haury sent A. E. Johnson (1960, 1963) down to work
at the site of La Playa near Cerro de Trincheras. Johnson conducted limited excavations on two
of the mounds, some features, and two cerros de trincheras sites nearby. Johnson (1960:42)
tested eight circular stone structures on the cerros de trincheras sites. He interpreted the “traces
of dry-laid masonry walls” as the remnants of houses. No cultural material was found below
ground surface of any of these. His excavation as a whole, led him to place the Trincheras as a
Di Peso (1956) sent Hinton (1955) to the Altar Valley in 1954. Hinton’s survey resulted
in a preliminary site typology and a chronology for the area. Hinton’s Altar Valley sites have a
high quantity of Trincheras purple-on-red ceramics. Hinton dated these sites to A.D. 700-1200.
His work led him to conclude that the people who inhabited these sites were a separate group: the
Investigators conducted a few more surveys to place the Trincheras culture in relation to
the Hohokam (McGuire and Villalpando 1995: 9). Danson (1946) surveyed the Santa Cruz River
Valley which reaches into Trincheras territory. Noguera (1958) also surveyed in Sonora and
published a summary of the archaeology of the state. Hayden (1956; 1970) conducted a survey
along the central coast of Sonora and the Sierra Pinacate. Wasley (1968) and Bowen (1976a)
17
surveyed the area west of Trincheras to look for evidence of Hohokam migrations into the area.
Bowen (n.d.) synthesized the information from their survey as an overview of the Trincheras
Tradition. Pailes (1972) directed a several year project in the Rio Sonora Valley.
Various researchers continued to synthesize the archaeology of the northern Sonoran and
southern Arizona. Stacy (1974, 1977) published the results of the first study of Papaguerian
cerros de trincheras. Stacy’s research allowed her to create a typology of the sites and the
features that occur on them. Stacy defined five rock work features based on morphological
characteristics: walls, terraces, circles, and bedrock mortars. She went on to refine the typology
by considering functional differences among the types. She defines seven functional categories:
enclosing walls, dividing walls, straight terraces, semicircular terraces, circular structures, trails,
and bedrock mortars. Stacy interpreted enclosing walls as ramparts with gates and dividing walls
as wing walls. She described straight terraces and semicircular terraces as residential or work
related spaces. Circular structures are defined as houses and described as stone features that are
usually grouped, have a diameter of 1 to 2 meters with a wall height of 0.5 to 1 meter, and have
noticible entrances. She interpreted the trails as access routes and bedrock mortars as food-
processing devices.
Bowen (1976b) later reported on the archaeology of the Seri. INAH issued results of the
first seminar on Sonoran archaeology (Braniff and Felger 1976). In 1977, Doyel (1977)
discussed Trincheras ceramics from sites in Santa Cruz Valley in Southern Arizona. Reinhard
The creation of the INAH- Centro Regional Sonora has resulted in the recording of many
sites in Sonora. Braniff and Quijada (1978) published information on several scattered sites in
18
Sonora. Braniff (1982) later presented the results of survey and excavation of Trincheras sites in
San Miguel River Valley and a summary of known archaeological sites in Sonora. Carrico (1983)
archaeological study at Cerro Calera. Elisa Villalpando (1985) reported on the study she
conducted at San Esteban Island. Braniff (1985) summarized much of the work INAH has done
in northern Sonora focusing on the Rio San Miguel and the Cerro Calera.
Simultaneously, archaeologists on the Arizona side of the border have gathered more
information on the nature of cerros de trincheras in southern Arizona. Wilcox and Larson
(1979) published research on cerro de trincheras in Tucson Basin. Wilcox fleshed out the
warfare. Cerros de trincheras are manifestations of internal instability for the Hohokam. Paul
Fish, Suzanne Fish and Chris Downum (1984; Downum 1986) began research on a cerro de
trincheras site in the Tucson Basin: Linda Vista Hill . They concluded that the terraces on these
sites served agricultural and habitational purposes (Fish et al. 1984: 11-12). The site is a Classic
Period Hohokam site (Downum et al. 1986). Downum also conducted research at Cerro Prieto
that he has interpreted as a Hohokam village site (Downum 1993; O’Donovan 1997: 56).
Northern Mexico has recently seen an influx of investigators from both sides of the
border. This factor promises a richer understanding of a long neglected region. Paul Minnis and
Mike Whalen completed a large-scale systematic survey of the areas around Casas Grandes
(Minnis and Whalen 1995). The survey led to the excavation of a large Casas Grandes site in
Chihuahua (Minnis 1998). Jane Kelley (Kelley and Stewart 1992) continues her investigation in
central Chihuahua. Robert J. Hard and John Roney are studying a three thousand year old cerro
19
de trincheras site in northwestern Chihuahua (Hard and Roney 1998). Rafael Cruz and Robert
Leonard are working at the Casas Grandes site of Galeana in Chihuahua (Leonard 1998, personal
communication).
The past decade has also seen new research in the Trincheras area. In the early 1980s
Randall McGuire and Elisa Villalpando (1993) began preliminary research in Sonora. In 1984
they conducted an extensive automotive survey of the Magdalena, Concepción, and Altar River
drainages and documented several cerros de trincheras sites. McGuire and Villalpando (1993)
conducted a systematic survey in the valley of Altar in 1988. In 1991 the investigators created a
systematic map of the site of Cerro de Trincheras (McGuire et al. 1995). The mapping project
served as the groundwork for the excavation of the site. Maria O’Donovan (1997) used the data
gathered by the mapping project to analyze the organization of Cerro de Trincheras. In this work
O’Donovan introduced the role of the monumentality of the site in interregional interactions.
McGuire, Villalpando, Vargas, and Gallaga (in press) have used the data from the excavation of
Cerro de Trincheras to relate it to the Casas Grandes world. Lupita Gonzales Sanchez and John
Carpenter are conducting research at the site of La Playa nearby (Carpenter et al. 1997).
will serve as the basis for future research. Additionally, current research will “repeople” a
approximately the international border to the north, Desemboque to the south, The Rio San
20
Miguel to the east and the Gulf of California to the west. Investigators are currently in the
21
Figure 3. Map showing archaeological traditions of Sonora
22
process of building the culture history for the Trincheras tradition. To date, the comprehensive
chronology was proposed by Bowen (1976a; n.d.) Bowen’s chronology is based on Wasley’s
(1968) survey. Bowen divided the Trincheras area into four regions: (1) fluvial; (2) coastal; (3)
the mouth of the Concepción river; (4) all remaining areas. The fluvial region includes the Altar,
Bowen defined Phase 1 as late archaic (?-A.D. 200). This phase looks very much like the
San Pedro phase of Cochise tradition of southern Arizona. Phase 2 (A.D. 200- 800) was mostly
concentrated in the coastal and the interior regions. The phase 2 sites look like the San Pedro
Cochise with the addition of pottery. This phase is characterized by settled villages and
Bowen describes Phase 3 (A.D. 800-1300) as a trend toward larger villages, Trincheras
purple-on-brown and Trincheras polychrome ceramics. Bowen placed the site of La Playa during
this period. Phase three saw the manufacture of shell jewelry and interaction with the Hohokam.
Bowen found Phase 4 (AD 1300-1450) sites concentrated in the fluvial areas. Bowen
defined this period as one characterized by a lack of local production of painted ceramics. He
placed the building of cerros de trincheras in this phase. He also noted a shift to cremation
Bowen did not define a protohistoric phase. Hinton (1955) and Carrico (1983) coincide
in an Oquitoa Plain to an Oquitoa red-on-brown and Altar red ceramic sequence for the
protohistoric Trincheras tradition. Both researchers linked these ceramic types with the
protohistoric Pima.
The Altar Valley survey and Braniff’s (1985) work in the Rio San Miguel take issue with
23
certain aspects of Bowen’s chronology (McGuire and Villalpando 1995, 1993). The Altar valley
survey allowed the refinement of the chronology. McGuire and Villalpando placed the
appearance of cerros de trincheras earlier during Phase 3 (A.D. 1100-1300 and Phase 4 (1300-
1450). The Altar Valley does appear to have a marked change to an O’otam artifact pattern. In
contrast outside the Altar Valley, Braniff’s radio carbon dates place Trincheras purple-on-red
ceramics well into the sixteenth century. This suggests a continuity of a Trincheras artifact
pattern to contact. Furthermore, Cerro de Trincheras appears to demonstrate this continuity till
A.D. 1450.
24
CHAPTER 3:
EXCAVATION PROJECT:
The Cerro de Trincheras Excavation Project provides investigators with the opportunity
to do basic baseline research on the Trincheras Tradition. This chapter describes the project
including its history and logistics. Then, I go on to describe the site in detail. The description of
the site provides a brief outline of excavation areas at Cerro de Trincheras. The final section
focuses on the project methods. The methods section includes excavation, recording and analysis
The Cerro de Trincheras Excavation Project was conducted as a joint venture by the
Sonora, Mexico and the State University of New York at Binghamton. Elisa Villalpando
Canchola of the INAH, Centro Regional de Sonora and Dr. Randall McGuire of SUNY
Binghamton directed the research. The National Science Foundation (NSF Grant SBR930224)
funded the project. Field work was done during the spring of 1995 and 1996.
The Cerro de Trincheras Excavation Project sought to gain more knowledge of Northwest
Mexico's prehistoric Trincheras Tradition. This thesis is a part of that project. Archaeologists
have largely ignored the pre-Hispanic populations of Sonora and Chihuahua. The knowledge
25
gained through the Cerro de Trincheras Research Project is an entry point to address questions of
relationships in regional and inter-regional contexts. The excavation of the site allows these
questions to be situated in time. Furthermore, detailed maps of the area did not exist prior to this
research, so maps created through the use of aerial photography have helped address issues of
Much of the project’s research centers on the creation of a fundamental culture-history for
Trincheras tradition. This thesis represents the first formal characterization of feature types for
the Trincheras culture. This has obvious implications for the understanding of Cerro de
Trincheras and will contribute to a general understanding of the archaeology of northern Sonora
and to the Northwest/Southwest area in general. Architectural classifications, like this one
inform broader archaeological and anthropological issues such as site structure, regional
interactions, etc., given importance organization of space at the village level. Though my study
alone cannot address these overarching issues, it is a part of a larger collaborative effort
(McGuire, Villalpando, Vargas and Gallaga: in press; O’Donovan 1997; McGregor 1998; and
others) that will create a synthesis of the pre-Hispanic village of Cerro de Trincheras which will
Site Description
The site of Cerro de Trincheras (Son F:10:2) includes the largest cerro de trincheras
recorded to date as well as three smaller terraced hills. One of the smaller hills lies to the east,
one to the south, and another to the west. The three smaller hills stand no higher than fifty
meters and contain a handful of terraces each. The large hill itself extends over 100 hectares and
rises 150 meters from the Magdalena River Valley overlooking the modern day town of
26
Trincheras, Sonora (McGuire et al 1991, O’Donovan 1997:1). Undeniably, this site is visually
striking even at a distance. It commands an impressive view over the Magdalena River Valley.
The main terraced hill (from here on referred to as “the cerro”, Spanish for the hill) is
covered by more than 900 terraces, 271 circular stone structures, 57 quadrangular stone structures
and several specialized features (O’Donovan 1997:106-109) . The cerro’s longest axis runs east-
west. The majority of the terraces are found on the northern side of the hill. The northern side is
consequently called the front. The crest of the hill at its highest, has three peaks and is divided
by two saddles.
Excavation Areas
The crew dug thirteen excavation areas in 1995 and 1996. The investigators excavated
sections on the north side of the hill, on the west side, at the crest, and off the hill on the south
side of Cerro de Trincheras (Figure 4). The crest of the hill was designated as area A. The north
side was identified as area B. The crew designated the south side of Cerro de Trincheras as area
C. Our team called the west side of the hill area D. The area located off the hill on the southern
side was designated as area E. Excavation areas within these sections of the hill were designated
by the letter indicating the area and were numbered sequentially (B1, B2, etc.) and each had
separate grid. The excavation areas vary in size and character. In the sections that follow, I
Area A
This crest of the hill was designated as area “A”. The highest and westernmost peak has
been termed el Pico de los Zopilotes or the buzzards roost (O’Donovan 1997: 70). The western
most saddle has been called el Abra del oeste or the western saddle. El Abra del oeste is an open
27
Figure 4. Map Showing Excavation Areas at Cerro de Trincheras
28
area enclosed by a series if terraces that descend from the middle peak or Pico de en Medio and
the Pico de los Zopilotes (O’Donovan 1997: 73). Several small terraces are found on the eastern
slope of the Pico de en Medio. The Pico de en Medio descends into the second saddle el Abra
del Este. The Caracol and several other structures are located in this eastern saddle. El Abra del
Este is delimited by terraces on its southern side (O’Donovan 1997: 75). The easternmost peak
is the least accessible since it is surrounded by walls and a terrace and has few architectural
features. Therefore, the crew located few architectural features in this area (O’Donovan
1997:78).
At the crest of the hill, one excavation area was examined: area A1. A1 was has been
called La Plaza del Caracol. This excavation locus includes areas surrounding the feature of El
Caracol. Many circular stone structures are also found in this area. Very little soil was found in
Area B
Area B is located on the northern side of the hill. We dug 11 excavation areas in this
section of the hill. The descriptions of the excavation areas begin at the base of the hill, proceed
up the west end of the northern side of the hill and descend to the eastern end of the northern
face. Area “B” includes “La Cancha” or the court, the large specialized feature at its base
(designated as B5).
The two large terraces directly south of La Cancha make up area B6. It consists of a long
wide terrace overlooking a second terrace that run east-west. On the surface several features are
visible. On the western side of lower terrace a circular stone structure and a quadrangular stone
structure are found at the back of the terrace. Both these stone structures use the terrace wall as
29
their back wall. A platform was recorded to the north of the circular stone structure on the lower
terrace. Two other circular stone structures were recorded towards the middle and eastern end of
Excavation area B1 consists of a set of two terraces (terraces 347 and 351). Terrace 351
was built first and was more substantial. The walls of terrace 351 were up to one meter high.
Terrace 347 is less substantial (terrace walls were up 0.60 meters high and up to a meter thick at
the base). The terrace wall of terrace 347 is more of a pile of irregular cobbles rather than neatly
Excavation area B2 is about 200 meters northeast of B1 lower down the slope of the
“cerro”. Area B2 is a single terrace (terrace 370). The terrace wall for B2 is quite substantial
measuring up to a 1.50 meters high and 1.5 meters thick at its base.
The project directors designated two terraces (terraces 330 and 313) sightly north and up
slope from B1 as excavation area B3. These two terraces are next to each other but have a clear
break and distinct terrace walls (McGuire & Villalpando 1995: 79-80). Terrace 330 is a double-
coursed wall built directly on bedrock. Terrace 330 had a maximum height of 0.65 meters and a
maximum thickness of 0.50 meters. Terrace 313 had two terrace walls with fill between these.
These walls were up to 1.2 meters high and 2 meters thick (McGuire & Villalpando 1995: 79).
The investigators defined terrace 329 as excavation area B4. Area B4 is located south
and up slope of excavation area B3. The terrace wall in B4 was more substantial than those in
areas B1 and B2. The terrace wall stands up to 1.25 meters and 2 meters thick.
The crew named the next excavation area up slope from B1, B7. Two terraces were
included in B7, Terrace 210 and 211. The area excavated measured 5 m wide by 27 m long.
30
Terrace 210 is a larger terrace that was constructed in front of Terrace 210.
The highest set of terraces (B8, B9, and B10) excavated were called el Mirador or the
lookout. These terraces have massive terrace walls sometimes exceeding four meters in height.
The configuration, its location, wealth of specialized materials (e.g. exotic ceramics), apparently
restricted access suggest that this area might have served as a household for the rulers of the site
(McGuire 1998: 4). The two terraces are connected by ramps that allow access from one terrace
to the next (B8 and B9). The second terrace, B8, has an ancillary terrace at its back southeastern
corner and a massive ramp which rises over three meters high. The next area up slope is B9. B9
is a terrace with walls. It has wall that wraps around the western, and eastern limits. These
walls left most of the southern and northern ends open. B10, the uppermost terrace has very little
B11 is located on the eastern side of the northern face of the hill. The excavation area
measured 9 meters by 24 meters. Three terraces were included in B11, Terrace 553 and Terrace
554. Terrace 554 is a small terrace behind Terrace 553. The terrace wall of Terrace 553 has a
Area D
McGuire and Villalpando designated the terraces at the base of the western side of the
cerro as area D. The crew excavated two loci in area D. Several stone structures that had been
defined during the mapping project were excavated as well as portions of terraces 3, 9, 11, 13,
15, and 17. McGuire and Villalpando (1995: 83) described terraces 3, 9, and 11 as wide (2 to 4
meters), long (10 to 19 meters) and low (0.50 to 0.75 meters). The researchers classified terraces
11, 13, and 17 as “narrow terraces”. Each of these had terrace walls formed by a single course of
31
medium cobbles and did not exceed 2.3 meters in width.
Area E
Excavation area E is located 500 meters south of the base of the hill. Since this area is
off the hill itself it was not built with terraces. Area E represents a very different zone since
constructions were built on fine sand instead of terraces. The highly porous sediment in area E
probably contributed to the bad preservation of features. Nevertheless, excavators recorded the
highest number of features with the most variability in this excavation area.
The nature of the excavation areas directly influenced the preservation and construction
of features at the site. All the areas the crew excavated occurred in one of the following sections:
the crest of the hill (A1), on the terraces (B1 thru B11 and D), or off the hill (E). Each of these
areas presents a distinct micro environment with differing effects on the deterioration and
construction of features. The top of the hill had very little sediment (usually less than 10 cm)
laying directly over bedrock. The excavators recovered almost all materials at ground surface
aiding to their deterioration. The cobbles readily available on the hill provided good building
material for the many stone structures at the crest. Additionally, the lack of sediment means that
all structures at the top of the hill were built directly on bedrock. The terraces represent a distinct
micro environment because they are man made constructions upon which other features were
built. The terraces at Cerro de Trincheras have a thin covering of very fine sediment
(approximately 15 cm). The shallowness and fine nature of the soil adversely affected on the
preservation of features on the site. Furthermore, this factor has made it especially difficult to
detect use surfaces of features and structures. The availability of stone, as on the crest, resulted
32
in the building of many stone structures on the terraces and access features between them.
Excavations off the hill found the highest concentration of features. In this case the micro
environment provided much more sediment which probably helped to preserve some features
Methods
The crew of the Cerro de Trincheras Research Project followed the methodological
guidelines in the field manuals from both the 1995 and 1997 excavation seasons (McGuire n.d.).
architectural features, as well, as procedures not included the manual. The archaeologists
followed the guidelines set forth by the excavation manual whenever possible. In all other cases
the archaeologists consulted the field supervisor, Randall McGuire to keep field methods as
standard as possible.
The excavation sample was extracted from the feature types and areas defined in the
Cerro de Trincheras Mapping Project (O’Donovan 1997). The mapping project defined several
units based on the concentration of surface artifacts on the hill. O’Donovan identified the
northern face of the hill as a residential area. She interpreted the crest of Cerro de Trincheras as a
ritual or ceremonial area. The size of the narrow terraces suggests they were not used for
habitation. These terraces are clustered at the northern and western bases of the hill. O’Donovan
interpreted these narrow terraces as agricultural. Finally, she describes La Cancha as a space for
public performance. The excavation was conducted to learn more about the nature of these units.
Provenience Control
Provenience control is essential when excavating any archaeological site For that reason,
33
the excavators followed very explicit guidelines in order to assure rigorous provinience control in
all excavation areas during both 1995 and 1996 excavation seasons. This was especially
The crew established a base line to create a grid and an elevation datum in every
excavation area (B1, B2, etc..). Geo- Map, a mapping firm from Tucson, mapped in the datum
points and grids on the general site map, establishing both the site grid and elevations. The base
lines in excavation areas ran from true North to true South on flat areas. In highly sloped areas, it
followed the long axis of the terrace being excavated. The elevation datum is a point on the
baseline. A floating datum of 100 meters was used to eliminate negative elevations and decrease
the occurrence of mathematical errors. The crew was instructed to record each level on a level
form and each unique datum or sub-datum has its own level form.
When the archaeologists encountered features during excavation, each feature was
assigned its own feature number. When the excavators located a feature within a feature, the
inside feature was called a sub-feature and designated by the primary feature followed by a
decimal point and its own sub feature number (e.g. 9.1). The field numbers were used in the
field and during analysis. The laboratory director issued the feature numbers in order to ensure
that the features were being numbered sequentially across the site. Once field work was
completed, the features were classified by types and new feature numbers were assigned (e.g.
pithouse 1) in order to ease comparison and write-up. A list feature numbers used in field and
The classification of features was finalized after excavation. The criteria used for the
typology was feature morphology, building method and associated sub-features. Once general
34
categories were established a subsequent analysis looked for information that would speak to
form and function. The information recorded about features in the field was summarized for
each feature type. This information was later used to compare Trincheras architectural with other
Excavation Procedures
The excavation of Cerro de Trincheras was aimed at furthering the understanding of the
large scale activity patterning at the site. Maria O’Donovan (1997) used the artifact and
patterning of activities at the site. Excavation areas were then selected within these activity units
at the site in order to better understand the activity units defined by the mapping project.
McGuire and Villalpando designed the excavation strategy to recover information that will be
used to make inferences about population, labor investment, chronology and activity structure.
The crew employed the following general excavation procedures both field seasons. Due
to the variation present on the site, no single excavation strategy was employed throughout.
Nevertheless the crew followed certain guidelines in all excavation areas. The laborers
performed all excavations by hand in 1 by 1 meter units. On the terraces the workers excavated
by brush and trowel. In area E, the only excavation area completely off the hill, most of the units
were excavated by shovel or trowel. Stratigraphic profiles were drawn whenever appropriate
collected artifacts in natural levels when possible. All fill with the exception of exploratory
trenches was screened through a 1/4 in or 1 cm screen to increase artifact recovery. Finer mesh
screens were occasionally used to recover smaller artifacts, to sample room floors and in smaller
35
feature fill.
In most cases when archaeologists detected buried features, these were excavated,
collected, mapped, and photographed in their entirety. The crew excavated all features by
natural or cultural levels whenever possible. All features were excavated by trowel and brush.
All artifacts associated with a feature were labeled with grid square, level and feature number.
Archaeologists completed sample forms whenever pollen, C-14, and flotation samples were
taken to record as much information on the association of the sample with the feature. All
samples were point provinienced. Excavation areas and features were plotted on the site map
When the workers detected a feature, buried or visible at the surface, a feature number
was assigned. In most cases a rough sketch map was drawn of feature prior to excavation. The
crew took photos prior to and after excavation in most cases. After analysis all features were
classified by type and assigned a corresponding feature number (e.g., Pithouse 1).
The archaeologists completed a feature form for each feature encountered. The
excavators recorded the date of excavation and the initials of the archaeologist supervising the
excavation of each feature. Each feature form (see Appendix 2 for forms used in field) called
for the excavation area feature located in, the feature number assigned, and the squares it was
found in. Archaeologists also recorded general descriptive measurements: maximum length,
width and depth on these forms. The starting elevation and level of origin for each feature was
recorded on each feature form as well. Feature number and description of photograph were
recorded in the photo log and roll numbers and frames numbers were recorded on each feature
36
form. The feature form also called for the relationship of the feature with other nearby features.
The crew recorded the feature numbers of all 1 by 1 meter units excavated in or including a
feature, on the level forms to facilitate cross-checks. Archaeologists wrote a description for each
feature recorded on the feature forms. Each feature form included a detailed map to scale of the
feature in plan view and in profile. All maps included a scale and a north arrow.
journals. In some cases, I referred to field journals in order supplement the information provided
Classification of features
After the conclusion of all field work at Cerro de Trincheras, I constructed a feature
Typology. The typology of features was constructed by using the feature forms and field notes.
The classification was made primarily on the basis of morphology and building strategy. In the
case of the pits, morphology and feature contents were used to classify the features. The specific
criteria used in my classification is explained for each feature type in Chapter 4. My typology is
Summary
The Cerro de Trincheras Research Project provides a first glimpse at a little known
cultural tradition. McGuire and Villalpando conducted excavations at the site in 1995 and 1996.
The sites monumental character is echoed in its internal organization. The project used
organizational units identified by the Cerro de Trincheras Mapping Project to guide the selection
of excavation areas. Thirteen excavation areas were dug. Each area has its own unique character
and internal organization. The complex nature of the site required the employment of very
37
explicit excavation methods. The excavation methods were outlined in the Cerro de Trincheras
Field Manual (McGuire n.d.) and are summarized in this chapter. The procedure for the
excavation, recording and subsequent classification of features also included in this chapter.
This chapter is intended to provide the spatial and methodological context for our excavation of
Cerro de Trincheras.
38
CHAPTER 4: FEATURE DESCRIPTIONS
The Cerro de Trincheras Research Project was the first large-scale excavation of a
Trincheras tradition site. The crew recorded and excavated 124 features on the site of Cerro de
Trincheras during the 1995 and 1996 field seasons. This chapter presents the results of the
excavation of the features by feature type. Each feature description includes descriptive data on
each feature excavated. First, I present a brief summary of the distribution of the features by
excavation area. Second, I describe the 4 terrace categories defined by the Cerro de Trincheras
Mapping Project. Next, I define the each of the general feature types found at the site: circular
stone structure, quadrangular stone structure, pithouse, “jacal”, pit, midden, platform, “ramada”,
bedrock engraving, rock arrangement, occupational surfaces, access features and burials. Each
feature type description includes summary data from each feature excavated. In the final section
of this chapter, I offer a summary and discussion of the data presented herein..
The crew excavated ten different terraces (B1-4 and B6-11), “La Cancha”(the court) (B5),
the top of the hill (A1), an area on the western side of the hill near the base (D) and an area off
the “cerro” on the southern side (E). Table 2 shows the distribution of the feature types by
excavation area with the exception of the three specialized features. The majority of the features
were located in excavation Area E (n=50). The archaeologists also found the largest variation
among features in Area E. Area A1 contains 17 features. Sixteen of these are circular stone
structures. The “Caracol” (the snail) is also located in A1. The crew recorded fourteen features
39
Table 2: Feature Type Distribution by Area/
Distribución de tipos de elemento por area
Featur e Type / To-
Tipo de elemento A1 B1 B2 B3 B4 B6 B7 B8 B9 B1 B11 D E tal
0
Access Feature/ 1 0 0 1 1 0 0 1 0 0 0 0 0 4
Elemento de ac ceso
Ancillary terrace/ 0 0 0 1 0 0 0 0 0 0 0 0 0 1
Balcon de terraza
Burials/ 0 0 1 0 1 4 0 0 1 0 1 1 2 11
Entierros
Pits/ 0 0 1 0 3 2 1 0 0 0 1 6 28 42
Hoyos
Jacal 0 2 1 0 1 1 1 0 1 0 0 0 0 7
Midd en/ 0 0 0 0 0 0 0 0 0 0 0 0 2 2
Basurero
Occup . Surface/ 0 0 0 0 0 1 0 0 0 0 0 0 3 4
Superfice
Ocupacional
Petrograbado 0 0 0 1 0 0 0 0 0 0 0 0 0 1
Pithouse/ 0 0 0 0 0 0 0 0 0 0 0 0 9 9
Casa sem i-
subterranea
Platform/ 0 0 0 0 1 2 0 0 0 0 0 0 0 3
Plataforma
Poss. Ramada 0 0 0 0 0 0 0 0 0 0 0 0 1 1
Rock A rrgn./ 0 0 0 0 0 0 0 0 0 0 1 2 3 6
Config. de piedras
40
in excavation area B6. Excavation area B6 is a few meters south of “La Cancha” (the court).
The team described twelve features in area D. The majority of which were pits. The
investigators recorded a total of 24 features in the remainder of the excavation areas. All of these
Terraces
The terraces are unarguably the most striking architectural features at the site. The labor
investment in the construction of the terraces was massive. The crew recorded eight-hundred and
eighty terraces recorded during the Cerro de Trincheras Mapping Project (see 0'Donovan 1997
for further description). During the excavation of the site 12 more terraces were recorded. These
features are what makes the site type unique. A feature description of the site would be
incomplete without a brief description of types of terraces present on the site. Archaeologists
defined three distinct terrace types were at the site: (1)terraces, (2) terraces with walls (Figure 5),
and (3) narrow terraces. Additionally, a complementary terrace type was identified and called an
ancillary terrace. Investigators established distinctions between the types on the basis of wall
Terraces
The type “terrace” was defined as those terraces that did not appear to have walls (Figure
6). Terraces were constructed by using dry-laid masonry of irregular unshaped cobbles to create
a berm like barrier which served as the terrace limit. The space between the natural slope of the
hill and the terrace wall was then filled with cobbles and trash in order to create a flat surface.
The terraces measured from 10 m to 100 m in length and averaged 21.02 m long (O”Donovan
1997: 101). The width ranged from 1.5 m and 5 m and averaged 3.54 m. The height ranged
41
from 0.30
42
Figure 5. A view of the terraces on the north face of the hill
O’Donovan found that the terraces with walls are the largest terraces at the site. Builders
constructed them using dry-laid masonry to form terrace walls. The terraces with walls were
constructed of core and veneer. In this method terraces are built by constructing two coursed
walls that are later filled with cobbles and trash. Then a flat surface was created by covering the
terrace fill with a thin layer of soil . The walls “were almost exclusively used to block access or
sight, into the higher portions of the site” (O’Donovan 1997: 105). Wall fall associated with
these structure suggests they once stood higher than they do today. Recorders found terraces
with walls mostly in the higher elevations. According to O’Donovan (1997:102), the slope of the
hill is such that it necessitates a greater investment in terrace wall construction to support the
weight of the fill necessary to create a flat surface. This is the case both in terms of height and of
wall thickness. The length of terraces with walls ranges from 10 m to 100 m and averaged 21.15
m. The width of terraces with walls ranges from 1.5 m to 5 meters, and averaged 3.5 m wide.
The height of the terraces ranges from 1.5 m to 3 m. As in the case of the terraces, the features
Narrow Terraces
Narrow terraces are the most insubstantial of the typology. Narrow terraces were
constructed by using a single course of irregular cobbles to mark its limits. The Mapping Project
recorded the narrow terraces in clusters at lower elevations of the site. The narrow terraces
measure on average 10.07 m in length, 1.95 m in width and 0.38 m in height. Due to their small
44
size, and clustering it is believed that these probably were single-function terraces (O’Donovan
1997:103).
Ancillary Terraces
Additionally, a terrace type that occurs on other terraces was also identified. These
complementary terraces were called ancillary terraces (Figure 7). Ancillary terraces occur on
some terraces These small complementary terraces were built on the surfaces of terraces with
walls. Limits of ancillary terraces look like sloping piles of cobbles. Therefore, ancillary
terraces do not have upright sides. Ancillary terraces range from 7 to 11 m in length and 2 to 3 m
in width. Up to twenty of these features were recorded in the higher parts of the hill (O’Donovan
1997: 104).
Cerro de Trincheras was a dynamic site on which its inhabitants carried out a wide range
of activities. The variety of features on the site stand as vestiges of once dynamic village. I have
established 13 general categories for features at the site. In the section that follows, I define each
feature type and describe the features the crew excavated during our field seasons.
Stone structures
Stone structures were found on the came in two configurations: circular and
quadrangular. The inhabitants of Cerro de Trincheras built stone structures with dry-laid
masonry using unshaped locally acquired andesite cobbles. The stone walls provided a base for a
perishable super structure. Three wall types were used in the stone structures. Some structures
combined these wall types. The Type 1, are stone foundations constructed of dry-laid or piled
unshaped irregular cobbles up in order to form a wall. Most walls of the stone structures were
45
built with
46
Figure 7. Ancillary Terrace
47
Type 1 walls. Type 2 walls are single coursed rock alignments along the perimeter of the
structure to serve as a base for a perishable superstructure. The Type 3 walls are terrace walls
incorporated into the perimeter of the structure. On average, the wall of stone structures stand 1
meter in height and typically 0.5 meter in width. In most cases these walls showed signs of some
wall collapse at the time of excavation. In the section that follows I will elaborate on each of the
stone structure types and characterize those excavated during the 1995 and 1996 field seasons.
Circular stone structures (Figures 8 and 9) are defined as round to oval structures
constructed of unshaped irregular sized andesite cobbles arranged in such a way as to provide
low walls upon which a perishable superstructure was constructed. For all circular stone
structures either stone foundations or terrace walls were used to mark the perimeter of the
structure. Circular stone structures have unprepared floors. In a few cases excavators detected
multiple use surfaces. The entries, when present, were built as a few flat laying cobbles, a break
in the wall, or a section of the perimeter wall standing much shorter than the rest. The team
recorded 271 of these structures during the Cerro de Trincheras Mapping Project (O’Donovan
1997: 105). This makes circular stone structures the most numerous features on the hill after the
terraces themselves. Circular stone structures appear to have served as domestic structures.
The crew excavated 21 circular stone structures in 1995 and 1996 (see Table 2). They
located one of these features inside the specialized feature, La Cancha. The circular stone
structures we excavated range from 6.2 meters to 1.9 meters in length with an average length of
3.90 meters. The width of the circular stone structures ranged from 1.5 meters to 5.6 meters with
48
an average
49
Figure 8. Circular Stone Structure 18
50
Table 3. Descriptive Statistics for Circular Stone Structures/
Datos Descriptivos para estructuras circulares de piedra
Number Provenience/ Length Width/ Walls Floor Entry/ orientation/
/Númer Area Procedencia /Largo Ancho Height Width / Entrada orientación
o North/ East/ (m) (m) (m) (m) Piso
Norte Este
51
The floors of the circular stone structures were generally difficult to discern. In all cases
floors are unprepared. For the most part the floors of the structures were identifiable by a slight
texture or color change. In a few cases, a layer of small cobbles and soil were arranged in order
to create a level floor surface. In one case, Circular Stone Structure 17 the floor was defined by
artifacts. In Circular Stone structure 18 at least two unprepared occupational surfaces were
detected as changes in consistency of the fill and the original floor level was constructed by
utilizing small cobbles and soil to level off the incline of caliche upon which the terrace was
built. Two floors were detected in Circular Stone Structure 19 as slightly harder packed surfaces.
The floor of Circular Stone Structure 20 was built upon a level of small rocks and soil 0.35
meters deep.
Large irregular, unshaped cobbles were used at the bases of the walls upon and around
which smaller rocks were arranged to roughly build up the side walls. The stone portion of the
walls appears to have served as a base for the remainder of the wall which is presumed to have
been built of perishable material. No obvious post holes or roof supports were found on these
stone walls suggesting lightweight building materials (e.g. brush) must have been used for the
remainder of the circular stone structures’ walls. The height of the walls ranged from 0.2
meters to 1.1 meters and averaged 0.66 meters. The width of the walls ranged from 0.2 meters to
The excavators found entryways in 80 percent of the circular stone structures. Circular
Stone Structures 18 and 19 had sloped ramp entries. Circular Stone Structure 20 did not show
signs of having a clearly marked entrance. The size of the entrances, when recorded, ranged
52
Evidence for the actual nature of the roofing of the Circular Stone Structures is limited.
There were no postholes or any other specialized roof supports found in association with either
the walls or the floors of any the excavated circular stone structures.
Recorders did not detect any intramural sub-features in any of the excavated circular
stone structures. Circular Stone Structure number four did have a small ash lens in the center of
its occupational surface. Yet, no formal hearths were found in association with these structures.
Excavators found a high density of artifacts inside structures of this feature type. Table 4
summarizes artifact distribution by circular stone structure. Plain ceramics and lithics were very
commonly recovered in these features. On average more than 800 pieces of plain ceramics and
more than 100 lithics were recovered from the interior of circular stone structures. Interestingly,
despite the low density of shell in most, circular stone structures 18, 19 and 22 all have more than
35 pieces each. Domestic material was highest in circular stone structures outside excavation
area A-1 (an average of 2833 plain ceramics and 527 lithics per circular stone structure outside
A1).
Quadrangular stone structures (Figures 10 and 11) are mostly rectangular in shape. These
structures are much larger than the circular stone structures. The mapping project recorded 57
quadrangular stone structures at Cerro de Trincheras. The builders used unshaped locally
acquired andesite cobbles. The occupants piled the cobbles using without mortar to build
substantial foundations along a four sided perimeter. All three wall types: stone foundations,
stone wall alignments, and terrace walls were used for walls in quadrangular stone structures.
Morphology, distribution and contents suggest these structures were used for habitation.
53
We
excavate
d four
quadran
gular
stone
structure
54
The crew excavated one quadrangular stone structure each in excavation areas B9, and B6, and
two in area D. The stone foundation walls of the quadrangular stone structures ranged from 0.3
meters to 1 meter in height. It is assumed that the remainder of the stone foundation walls if the
quadrangular stone structures was constructed of perishable material using the short stone walls
as a base. Quadrangular Stone Structure 1 has stone foundations and terrace walls for walls.
Quadrangular Stone Structure 1 shares its northern wall with Circular Stone Structure 18.
Terrace 576's terrace wall serves as the back wall of Quadrangular Stone Structure. Since the
stones from the terrace wall 576 and Quadrangular Stone Structure 1 do not interlock, it suggests
that Quadrangular Stone Structure 1 was built after the construction of terrace 576.
Quadrangular Stone Structure 2 uses stone foundations and stone wall alignments. The northern
and eastern walls were definable by alignments of small to medium sized cobbles standing at a
maximum of 0.30 meters. In the case of the northern wall this was constructed directly above
the back wall of terrace 280. The excavators noted that the walls of Quadrangular Stone
Structures 3 and 4 were not well preserved by the amount of wall fall present at the time of
excavation.
The use surfaces of Quadrangular Stone Structures were difficult to discern. The floors
of Quadrangular Stone Structure 1 were not clearly definable. Several changes in color and
surface was built upon a level of terrace fill built up higher than the remainder of the terrace
surface. The crew did not locate an occupational surface for Quadrangular Stone Structure 2.
The use surface of Quadrangular Stone Structure 3 was only indicated by a level of small rocks.
Quadrangular Stone Structure 4 had a floor also defined by small rocks. Presumably, the rock
55
floor of Quadrangular Stone Structures 3 and 4 was cover with layer of soil to create a flat
56
Table 4. Distribution of Material by Circular Stone Structure/Distribución de material por
estructura circular de piedra.
Circular Stone Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Structure/ Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Concha
Estructura Ceramica Ceramica Herramie Litica
Circular de Lisa Decorada ntas de Pulida
Piedra Piedra
1 A1 365 0 5 0 0 5 1
2 A1 860 0 197 1 2 0 0
3 A1 0 0 10 0 0 0 0
5 A1 507 0 68 0 0 7 1
6 A1 0 0 5 0 0 0 0
7 A1 207 0 58 2 0 0 1
8 A1 552 0 102 1 1 4 6
9 A1 8 0 19 0 0 0 0
10 A1 60 0 8 0 0 0 0
11 A1 39 0 10 0 0 0 0
12 A1 17 0 6 0 1 0 1
13 A1 4 0 4 0 0 0 0
14 A1 106 0 3 2 4 0 0
15 A1 270 0 233 0 1 2 2
16 A1 42 0 9 0 1 0 0
17 B3 1436 0 110 0 4 54 2
18 B6 5093 9 1241 15 9 32 46
19 B6 4876 16 862 18 7 13 46
20 B6 2187 0 304 10 3 21 12
21 A1 51 0 0 0 0 0 0
22 A1 3836 0 799 5 3 13 39
23 A1 5 0 141 1 0 0 0
24 A1 462 0 129 0 1 4 2
57
surface.
Table 5. Descriptive Statistics for Quadrangular Stone Structures/ Datos descrptivos de estructuras
cuadrangualares de piedra.
Quadangular Provenience Wall
stone structure/ Area N E Length Width Height entry orientation
Estruct. Cuadr. (m) (m) (m)
de piedra.
the structures. The archaeologists located the entryway to Quadrangular Stone Structure 1 at the
North East corner measuring 1.15 meters wide. The entrance of Quadrangular Stone Structure 2
was not found due to frequent breaks in the alignments that makeup type 2 walls. Presumably
the entrance faced east because excavators found no breaks in the remainder of the Quadrangular
Stone Structure 2. Excavators defined two entrances for Quadrangular Stone Structure 3. The
entrance to the Southeast had a 4 meter long rock alignment parallel to the western wall that
formed a hallway into the feature. The northeastern entrance lead to Terrace 9 above the
structure.
The crew found little evidence for the nature of roofing of the quadrangular stone
structures. No postholes or any other sub-features that could have served as roof support or wall
support. This suggests the superstructures was made of perishable material. No hearths, or any
other intramural sub-features were identified during the excavation of the Quadrangular Stone
58
Table 6. Distribution of Material by Quadrangular Stone Structure/ Distribución de material por
estructura cuadrangular de piedra
Quadrangular Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Stone Structure/ Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Estructura Ceramica Ceramica Herra Litica a
Cuadrangular de Lisa Decorada mientas Pulida
Piedra de
Piedra
2 B9 6167 0 314 0 1 8 12
3 D 6129 13 532 7 4 68 80
4 D 1732 0 492 6 4 14 59
Structures.
The excavators recovered a high quantity of artifacts from the interior of Quadrangular
Stone Structures. Table 6 summarizes the counts for each material type by quadrangular stone
structure. Shell, plain ceramics and lithics were commonly found in these structures.
Pithouses
The pithouses (Figures 11 and 12) at Cerro de Trincheras conform to the basic
understanding of the term house in pit. Houses in pits are defined as structures built in
depressions where side walls rise to meet the roof (Haury 1976: 71). True pithouses used edges
of pit as a side walls. Trincheras pithouses have ramped entries. At Cerro de Trincheras the
depressions pithouses were built in were no deeper than a few centimeters at the time of
59
Figure 10. Photo of Pithouse 1
60
Figure 12. Pithouse 9 Ramp
61
excavation. Trincheras pithouses probably served as habitation structures
The archaeologists identified nine pithouses at Cerro de Trincheras. All pithouses were
located in area E. Excavators defined Pithouses 6, 7, 8, and 9 by ramp entries (Figure 12) and
their spatial distribution. Their corresponding pithouse floors did not survive. Table 7
summarizes descriptive statistics on the pithouses excavated. The pithouses range from 2.57
meters to 4.35 meters and averaged 3.46 meters in length. The width of the pithouses ranged
from 2.25 meters to 3.50 meters and averaged 3.21 meters. The floors of the pithouses were not
uniform. The pithouses were dug into native caliche which was manipulated in order to create a
flat occupational surface. The outcropping caliche resembled a burnt, adobe plastered, surface in
some areas. Pithouse 4 has a portion in its NE corner that appears to have been burned. Since
Pithouses 5-9 were identified after excavation on information was recorded about their floors.
Access into pithouses was provided by ramps entries. The ramps were constructed of
heavily burned caliche creating a step on one end and a ramp on the other. Recorders did not
The excavators recorded no postholes to support for roofing. The crew reported rock
rings along the perimeter on a level above the pithouse use surfaces. These rock rings were most
probably helped support the walls of structure. The absence of substantial roof support suggests
that the roofs of these structures were constructed of light perishable materials.
A hearth was located in Pithouse 2. The hearth’s boundaries were determined by edges
of burnt caliche lined with a few rocks that cut into feature fill. It southernmost edges showed
signs of being especially burned. This hearth was filled with ash, as well as a small quantity of
62
Table 7. Descriptive Statistics for Pithouse/ Datos Descriptivos de “pithouses”.
ceramic and lithic material. A possible hearth was located inside Pithouse 4. This possible
hearth was identified as a small oval area of burnt caliche with a slight ridge only a few
centimeters high. This possible hearth was located in the northeastern corner of Pithouse 4.
Excavators relatively little material inside pithouses (Table 8). The plain ceramics
(<1000
except for Pithouse 1) and lithics (<200) make up the largest material type in the pithouses.
63
Since the exact perimeter of Pithouses 6 thru 9 was never determined artifacts could not
confidently
64
Table 8. Distribution of Material by Pithouse/ Distribución de material por “pithouse”
Pithouse Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Ceramica Ceramica Herra Litica a
Lisa Decorada mientas Pulida
de
Piedra
1 E 1122 0 198 1 3 76 0
2 E 422 0 183 5 2 1 2
3 E 263 0 164 0 2 43 1
4 E 27 0 16 0 3 0 0
5 E 1 0 1 0 0 0 0
6 E 0 0 0 0 0 0 0
7 E 0 0 0 0 0 0 0
8 E 0 0 0 0 0 0 0
9 E 0 0 0 0 0 0 0
assigned. The crew found only recorded three shell artifacts from inside the pithouses.
Jacal Structures
The term “jacal” refers to wattle and daub houses. These could of been constructed of
brush, sticks, and ocotillo often covered with mixture of mud. Upon excavation jacales appear
as medium and small irregular cobbles arranged along a generally oval perimeter. These rocks
formed part of the structure walls by providing support for walls, of perishable materials or were
simply cleared from the space jacales were built in. We use the term “Jacal” with the
understanding that all of our structures (including our stone and pithouse structures) could
technically be termed “jacal” structures. Yet, we reserve the use of the term “jacal” to those
65
Figure 13. Jacal structure
66
Table 9. Descriptive Statistics for “Jacal” Structures/ Datos Descriptivos de jacales
Jacal Area Provenience Length/ Width Depth Orientation/
Proced encia Largo Ancho Profundidad Orientación
N E (m) (m) (m)
The crew identified eight “jacal” structures during the excavation of Cerro de Trincheras.
Table 9 presents descriptive statistics for the “jacales” excavated. Two jacales were found in
area B1, one in B2, one in B4, one in B6, one in B7 and one in B9. The lengths of the “jacal”
structures range from 2.26 meters to 6.5 meters and average 4.74 meters. The width of the
“jacales” ranged from 2.20 meters to 5.13 meters and averaged 3.77 meters. The depth of these
The archaeologists had a difficulty defining the floors of the “jacal” structures. Very
slight discontinuous changes in texture and concentartions of artifacts were observed in several
67
of these structures and were recorded as possible use surfaces. In the case of “Jacal” 4 native
caliche was used as part of the original floor surface while the remainder was created by placing
small rocks and soil in order to create a flat surface. These structures appear to have multiple
occupational levels which do not manifest themselves evenly throughout the extent of the
“jacales”.
The crew had difficulty finding the exact location and nature of entrances into the jacal
structures due to the lack of contiguous wall alignments. No entrances were located in “Jacales”
1, 2, 3, 5, and 8. Entrances were noted in three of the jacal structures. A possible entrance was
located on the NW extreme of Jacal 4.This entrance was defined by a gap in the alignments that
constitute the walls of the structure. The entrance into Jacal # 6 was located as a gap in wall
supporting cobbles on the NE side of the jacal. Two entrances were found in association with
Jacal 7. The SE entrance of Jacal 7 has a line of rocks which form a type a small hallway
limited by the entrance on its east end and the west wall. The NE entrance of Jacal 7 marked by
The excavators found evidence for roofing Jacales 3 and 4. Seven postholes were found
in association with Jacal 3. Four of the seven postholes were deep and well defined. Jacal 3 also
yielded several fragments of daub some of which showed ocotillo impressions. Several
fragments of burnt daub as well as one with well-defined ocotillo impressions were collected
from within Jacal structure 4. The remainder of jacal structures showed no traces of roofing.
Excavators only recorded one hearth inside the “jacal” structures. A small hearth (0.30
meters diameter) was found in association with Jacal 4. The walls of the hearth showed evidence
of burning. The hearth was filled with ash and small charcoal flecks. Jacal 4 was the only
68
feature of this type with a hearth in its interior.
The archaeologists found a buried ceramic jar associated with jacales. Pit 17 was a pit
with a ceramic vessel in it. Pit 17 was built directly outside the entrance of Jacal 4 on the eastern
end of the structure. The excavators located another buried ceramic vessel beneath the interior
occupational surface of Jacal 2. Charcoal flecks, a number of ceramic fragments and small rocks
The crew recovered a high quantity of domestic material from the interior of “jacales”.
More than 3000 plain ceramics and 200 lithics were recovered from each of the jacales
excavated. Vargas analyzed a total of 272 pieces of shell were analyzed from jacales. The
excavators collected more than 26,000 plain ceramic pieces from Jacal 5 alone. Jacal 7, located
in B7, had more than 51 percent (n=138) of the total shell found in the jacales. Table 9 shows
Platforms
The crew defined three platforms during the excavation of Cerro de Trincheras (Figure
14). I distinguish platforms from terraces on the basis of construction strategy. Platforms do not
havewalls to delimit them . These platforms are generally limited by an alignment of medium
sized cobbles at its extremes. Furthermore, platforms are built on the fronts of terraces. Piled
rocks were used to create a level surface higher than the terrace surfaces it was built upon. Rocks
and soil were used in order to create a level surface higher than the original terrace. These
The platforms varied greatly in size. The largest, Platform 1, measured 11.50 meters in
length and 6.60 meters in width with a maximum thickness of 0.49 meters. Platform 2 measured
69
Figure 14. Platform 2
70
Table 10: Distribution of Material by Jacal./ Distribución de material por jacal.
Jacal Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Ceramica Ceramica Herra Litica a
Lisa Decorada mientas Pulida
de
Piedra
1 B1 6672 7 662 6 1 7 17
3 B4 2669 10 228 13 10 16 16
4 B6 5920 7 1196 15 1 9 22
6 B9 9778 5 632 1 2 22 25
7 B1 3811 9 266 0 7 5 20
9.83 meters by 7.67 meters with a maximum thickness of 0.15 meters. The dimensions of
Platform 3 were 9.83 meters to 7.67 meters with a maximum thickness of 0.15 meters.
The floors of the platforms were built by creating a level surface of small rocks covered
by soil. These surfaces were detectable as a change in texture along a horizontal plane. The
occupational surfaces of Platform 1 and 2 were some of the most well defined surfaces on the
entire site. The floors of Platforms 1 and 2 were hard packed surfaces with archaeological
material laying on them. The workers found a high number of artifacts lying flat on the surface
recovered from the surface of Platform 2. The excavators did not locate an occupational surface
for Platform 3. Platforms 1 and 2 were built adjacent to Terrace 576. Platform 1 was built up
against and has access into Circular Stone structure 18. Circular stone structure 19 was built on
the western extreme of Platform 2 connected by a small ramp. Circular stone structure 20 was
71
constructed onto the north wall of Platform 2.
The boundaries of these platforms were discernable as alignments of rocks higher than
the level of the terrace it was constructed on. This difference in height manifested itself as a sort
of step. There is no evidence to suggest that these platforms were ever walled or roofed
A high quantity of artifactual material was found on the surfaces of platforms. More than
plain ceramics were collected from each of the platforms. The excavators recovered more than
500 lithics and 10 stone tools from each platform. The artifactual material further suggests that
Occupational Surfaces
unprepared surface (Figure 15) or a concentration of material along a horizontal level (Figure
16). The crew recorded four Occupational Surfaces (see Table 12 for descriptive statistics).
Three occupational surfaces were located in area E and one in area B6. Each occupational
surface is described in the section that follows. Artfacts collected in association with the
plainware). The highest concentration of which was located on its NW extreme. The crew did
72
Figure 15. Photo of Occupational Surface 3-
hard Packed Surface
73
Figure 16. Photo of Occupational Surface 1-
concentration of artifacts
74
Table 11. Distribution of Material by Platform. / Distribución de material por plataforma.
Platform/ Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Plataforma Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Ceramica Ceramica Herra Litica a
Lisa Decorada mientas Pulida
de
Piedra
1 B6 14290 9 3154 40 3 27 49
2 B6 15012 1 2568 51 16 41 86
3 B4 6526 10 541 11 2 75 40
Recorders describe it as a hard packed surface with several flat-lying artifacts upon it. This
surface was 0.18 meters thick. A zoo-morphic shell pendant, a shell bead, and some small burnt
animal bones were recovered from beneath this surface. This surface overlaid Hearth 8.
Therefore, the burnt bone might be associated with it. No shell was found in association with
Occupational Surface 2.
Occupational Surface 4 as a hard packed surface with flat laying artifacts on it. Occupational
surface 3 overlays Hearth 3. Vargas only recorded 2 pieces of shell on Occupational Surface
small rock platform and two rows of stones which could represent the bases of two parallel walls.
The hearth was roughly rectangular in shape with rock and adobe sides. The hearth has a
maximum width of 0.23 meters and a maximum depth of 0.16 meters. It was filled with grey
75
soil,
76
Table 12. Descriptive Statistics for Occupational Surfaces/ Datos Descriptivos de superfices
ocupacionales
Occupational Surface/ Area North/ East/ Length/ Width/
Superfice Ocupacional Norte Este Largo Ancho
(m) (m)
1 E 102-109 111-118 7 7
1 E 106 0 64 3 2 0 0
2 E 32 1 42 0 0 19 0
3 E 191 0 175 0 0 22 2
4 B6 3366 0 792 9 6 0 2
charcoal and ash. The small rock platform was made of approximately flat stones arranged on
the terrace surface. Two parallel rows of rocks set in adobe were located SW of the hearth. It is
possible, yet inconclusive, that these rows of rocks may have served as foundations for walls of
a perishable material. Artifacts were especially clustered in two areas to the southeast and to the
northwest of the hearth. These clusters contain large fragments of ceramic vessels and manos. A
77
fragment of a metate was also located on this surface.
Access Features
The large number of terraces on the slopes Cerro de Trincheras necessities access features
to lead from a terrace or area to another. In some cases terraces walls exceed 3 m in height. In
cases like this ramps are required to get from lower to upper terraces. Access features at the site
were identified in two forms: ramps and stairs. One ramp was located in A1, another in B8. The
crew recorded two sets of stairs one in area B4 and the other in B8.
Ramps
Two ramps were identified during the excavation of the site. These ramps were
constructed using the same method of construction for the terraces on. Rubble fill was piled to
create a uniform sloping surface that connects one terrace surface to another. The rocks are then
The excavators recorded Ramp 1 in area A1. Ramp 1 was built along the south wall of
terrace 716 (prov. N 85-90, E 276-280) leading to El Caracol. This ramp provides the southern
access to the plaza of area A1. It is 3.70 meters long and 1.70 meters wide with a maximum
Ramp 2 was located in area B8 (prov. N 97-100, E 276-283). Ramp 2 connects terrace
280 with 278. It is located on the center of the posterior of terrace 280. This ramp is 7.00 meters
long by 3.00 meters wide. It has a SW-NW orientation. On its northern extreme it is bounded by
the west back wall of terrace 280 which measures 0.21 meters to 0.67 meters wide measured
from the interior of the ramp and 1.93 meters from the exterior of the ramp. To the south the
ramp is limited by the main wall of terrace 277. From terrace 280 the ramp rises as a narrow
78
ramp that
79
Figure 17. Photo of Ramp 2
80
Table 14. Distribution of Material by Access Feature/ Distribución de material por elemento
de acceso
Access Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Feature/ Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Elemento de Ceramica Ceramica Herra mie Litica a
Acceso Lisa Decorada ntas de Pulida
Piedra
Ramp 1 A1 134 0 20 0 0 0 0
Ramp 2 B8 963 0 61 0 0 0 3
Stairs 1 B4 0 2 0 0 0 2 0
widens as it rises toward terrace 278. Very little artifactual material was collected from ramps.
Stairs
Two sets of stairs were recorded during excavation (Figure 18). The stairs are fashioned
of rock alignments with fill placed along the face of bedrock. These stairs provide an access
from one terrace to another. Very little material was collected from these features. Victoria
Vargas’ analysis recorded no shell artifacts were recovered from the stairs.
Stairs 1 (prov. N 106-108, E 120-123) in area B4 connects Terrace 329 with Terrace 330.
These stairs are approximately 2.5 meters wide. This feature consists of three steps. It is located
in the NE corner of terrace 329 providing the eastern boundary of the terrace. This feature was
Stairs 2 were located as access between Terrace 330 and Terrace 313 (prov. N 106-109,
E 120-124) in area B3. These stairs are not very well preserved. They measured 3.00 meters by
2.50 meters. These appear as rock alignments against bedrock using rock and soil in order to
81
Figure 18. Photo of Access Feature: Stairs 1
82
create a flat surface. These stairs are located to the east of Ancillary Terrace 1.
Pit Features
The excavators found pit features to be the most numerous feature category at the site
(n=43). Pit features were found in excavation areas B2, B4, B6, B7, B11, D, and E. At Cerro de
Trincheras, pits probably facilitated the realization of a variety of different activities. In the field
this general pit category was divided into three broad categories: pits, hearths, and puddling pits.
The section that follows describes the pits and attempts to further categorize them into broad
types on the basis of their contents and morphology. Table 14 summarizes the distribution of
Pits
Pits were defined as depressions without evidence of burnt walls (see Table 15 for
descriptive statistics of Pits). The crew recorded 24 pits at the site. After analysis, the category
“pit” was further broken down on the basis of shape and contents. Consequently, I have defined
nine pit types: multipurpose pits, caliche lined pits, trash filled pits, storage and processing pits,
roasting pits, ash pits, and undifferentiated pits. To standardize the typology I have employed the
criteria used in the Pueblo Grande Project by Mitchell and Merewether (1994: 85-105) in
Phoenix, AZ to classify the pits found at in Trincheras. The section that follows represents a
Multipurpose pits
The multipurpose pits were defined depressions of variable size and shape that show
evidence of at least two of the following activities: storage, processing, food preparation
83
. Pit Data/ Datos de hoyos
Pit #/ Loci/ North/ East/ Depth/ Length(diam.)/ Width/ Plan Fill/ Stratified
Numero Area Norte Este Profundidad Largo Ancho view/ Relleno deposits/
de Hoyo Plano Depositos
Estratigrafi-
cados
84
multiple
85
Table 15 continued Pit Data/ Datos de Hoyos ceniza no
128.00- 144.17- Ash pit/ Irregular/ Ash &
128.90 145.10 Hoyo de ceniza Iregular Carbon/
carbon
ceniza no
124.20- 128.60- Ash pit/ Oval/ Ash &
125.00 129.40 Hoyo de ceniza Ovalado Carbon/
carbon
ceniza no
109.25- 128.15- Ash pit/ Irregular/ Ash &
109.95 128.65 Hoyo de ceniza Iregular Carbon/
carbon
ceniza no
Pit #/ Loci/ North/ East/ Depth/ Length(diam.)/ Width/ Plan view/ Fill/
Numero Area Norte Este Profundidad Largo Ancho Plano Relleno
de Hoyo Stratified
deposits/
Depositos
Estratigrafi-
128.50- 139.50- Ash pit/ Irregular/ Ash &
129.30 139.95 Hoyo de ceniza Iregular Carbon/ cados
carbon
ceniza no
106.30- 114.00- Ash pit/ Round/ Gen.
106.50 114.90 Hoyo de ceniza Circular Cult./
Rell.
Gen. no
Cult
86
. Pit data/ Datos de125.15-
hoyos
127.00 120.75 Tipo no Circular Varios
identificado
no
123.60- 118.70- Unknown/ Oval/ Varied/
124.20 119.80 Tipo no Ovalado Varios
identificado
no
109.00- 321.00- Unknown/ Rectangula Varied/
110.00 322.00 Tipo no r Varios
identificado
no
86.60- 70.05- Unknown/ Oval/ Gen.
87.10 70.54 Tipo no Ovalado Cult./
identificado Rell.
Gen. no
Cult
87
88.20 70.35 Tipo no Ovalado Cult./
identificado Rell.
Gen. no
Cult
Table 15 continued. Pit Data/ Datos de Hoyos
89.00- 70.01- Unknown/ Round/ Gen.
89.50 70.69 Tipo no Circular Cult./
identificado Rell.
Gen. no
Cult
Table 16. Distribution of Material by Pit/ Distribución de material por hoyo
Pit/ Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Hoyo Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Concha
Ceramica Ceramica Herram- Litica
Lisa Decorada ientas de Pulida
Piedra
Pit 5-Multipurpose E 7 0 7 0 0 0 0
Pit 12-Caliche E 0 0 4 0 0 1 0
Lined
Pit 13-Unknown E 3 1 8 0 0 9 1
Pit 19-Unknown D 6 1 0 0 0 6 1
Pit 20-Unknown D 0 0 0 0 1 13 4
Pit 21-Unknown D 12 0 1 0 0 0 0
Pit 22-Unknown D 75 0 0 1 0 3 2
Pit 23-Unknown D 0 0 0 0 0 0 0
Pit 24-Roasting D 99 0 12 0 0 0 1
88
involved in construction, and trash deposit. Archaeologists excavated four multipurpose pits at
the site. All four of these pits occurred in excavation area E. Two of the multipurpose pits were
oval (pit 1 and pit 2). Pit 3 had an irregular shape and pit 5 was round. The dimensions of
multipurpose pits ranged from 0.44 meters to 2.40 meters in length and 0.50 meters to 2.05
meters. The multi functional pits range from 0.20 meters to 0.94 meters in depth. The average
depth of the multi functional pits is 0.42 meters. Only one of these multi functional pits (pit 2)
had stratified deposits. In pit 2, the excavators found a more recent level of trash deposits and
older level of ash and charcoal separated by a level of water laid deposits.
The crew recorded two caliche lined pits: (pit 4 and pit 12) The type was defined solely
on the presence of a caliche lining on the perimeter of these features (Figure 19). The fill of
these caliche lined pits differed. Pit 4 was filled with general cultural material, while pit 12 was
filled a mixture of fire-cracked rocks, ash, charcoal, and a small number of artifacts. Both
caliche lined pits occurred in excavation area E. Pit four was round with a diameter of 0.80
meters. Pit 12 was oval (0.70 by 0.45 meters). They had an average depth of 0.13 meters. These
The trash filled pit was defined a depression with well defined walls, filled with trash. I
differentiate this category from that of midden on the basis of the contents of the feature having
been found in the interior of a depression. The archaeologists recorded only one of these: pit 15.
Pit 15 is located in excavation area B4. It is round and measures 2.00 meters in diameter with a
maximum depth of 0.43 meters. It was filled with variety of material including ash, ceramics and
89
Figure 19. Photo of Pit 4 Caliche Lined Pit
90
Figure 20. Map of Pit 4: Caliche Lined Pit
91
rocks. Excavators did not find stratified deposits in its interior.
Mitchell and Merewether (1994: 85-105) defined the storage and processing pits as those
depressions that show clear evidence for storage and/or processing. Archaeologists recorded one
of these features at the site: pit 17. Pit 17 was an oval depression measuring 0.71 in length and
0.56 meters in width. This depression contained small undecorated vessel whose rim was placed
so as to appear flush with the living surface of the terrace it was built on. This feature was
located adjacent to “jacal” 14 and most probably served as a storage feature associated with it.
Roasting pits
Excavators identified two features as roasting pits (Figure 21). Roasting pits were
recognized as depressions with some or all of the following: ash and charcoal in the fill, in situ
burning, and fire cracked rock in the fill. In none of the roasting pits were the deposits stratified.
Pit 14 was recorded in excavation area B2 while pit 24 was located in area D. Pit 14 was round
and measured 1.32 meters in long, 1.12 meters in wide and 0.45 meters deep. Contrastingly, pit
24 was irregular in plan view and measured 3.44 meters long, 0.70 meters wide and 0.12 meters
deep.
Ash pits
Recorders identified six features as ash pits (Figure 22). These features were defined as
depressions filled with primarily ash, as well as small amounts of charcoal in the absence of
evidence of burnt walls. The lack of evidence for in situ burning indicated that these pits were
not the location of the primary processing activities which produced the ash. Therefore, it is
most probable that they served at receptacles for discarding ash. The ash pits were irregular in
92
plan.
93
Figure 21. Photo of Pit 14: Roasting Pit
94
Figure 22. Photo of Pit 8: Ash Filled Pit
95
(n=4). Pit 16 was round and pit 8 was oval. All but one of these pits was located in area E. Pit
16 was recorded in area B4. The ash pits depth ranged from 0.10 meters to 0.38 meters with a
mean depth of 0.17 meters. The ash pits’ length ranged from 0.95 meters to 0.50 meters with a
mean length of 0.72 meters. The width of the ash pits ranged from 0.42 meters to 0.86 meters
with an average width of 0.59. The crew did not find any ash pits with stratified deposits.
Undifferentiated pits
The majority of the pits recorded could not be classified because the did not have clear
evidence of a function. Undifferentiated pits were found to be unprepared. Eight of these were
found throughout the site. Five were recorded in area D, one in area B11, and two in area E.
They were found with rectangular(n=1), circular(n=3), and oval (n=4) plan views. None of their
contents were stratified. The depth of the undifferentiated pits ranged from 0.20 meters to 0.76
meters, with an average of 0.34 meters. Undifferentiated pits ranged from 1.58 meters in length
to 0.50 meters, with an average of 0.88 meters. The width of the undifferentiated pits ranged
from 1.43 meters to 0.40 meters, with a mean width of 0.66 meters.
Puddling pits
Puddling pits were defined as depressions with caliche lined sides and bottoms and
caliche in feature fill (Figure 23). Two circular caliche puddling pits were recorded at the site
(see Table 16 for descriptive statistics). Excavators found both of these features in area E. It is
differentiated from the caliche lined pit by the presence of caliche plaster in its interior. These
pits were presumably used for the mixing of caliche and soil for construction purposes. No shell
was collected from these structures. Caliche puddling pits had an average depth of 0.11 meters.
Recorders found the average diameter for these features was 0.43 meters. Pudding pit 1 is
96
Figure 23. Photo of Pudding Pit 2
97
Table 17. Descriptive Statistics for Pudding Pits/ Estadisticas descriptivas para pozos para
mezclar caliche.
Pud.P it/ Loci/ North/ East/ Depth/ Diameter/ Plan view/ Stratified deposits
Numero de Area Norte Este Profundidad Diametro Plano
Hoyo
Table 18. Distribution of Material by Pudding Pit/ Distribución de material por pozo para
mezclar caliche
Pudding Pit/ Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Pozo d e Me zcla Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Conch
Ceramica Ceramica Herra Litica a
Lisa Decorada mientas Pulida
de
Piedra
1 E 23 0 0 0 0 0 0
2 E 0 0 0 0 0 0 0
Total 23 0 0 0 0 0 0
associated with pithouse 5. Pudding pit 2 is near by and probably associated with pithouse 1.
Hearths
The crew defined hearths as those depressions whose walls showed evidence of in situ
burning. Excavators recorded 17 hearths at the site. Hearths were located in excavation areas
B4, B6, B7, D, and E. All but two of the hearths recorded at the site were prepared. The
98
depressions ranged from 0.08 meters to 0.63 meters deep. The dimensions of the hearths
recorded at the site measured from 0.33 meters to 1.30 meters in length and from 0.22 meters to
1.10 meters in width. The average dimensions of the hearths were 0.61 meters long, 0.50 meters
wide, and 0.20 meters deep. Table 19 summarizes descriptive statistics for the hearths we
excavated. The hearths are generally round (n=11) to oval (n=5) (Figure 24), with the exception
of hearth 5 which is a curved semi-circle and hearth 6 which recorders categorized as irregular in
plan view (Figure 25). The fill of these structures varied considerably including ash and
charcoal, general cultural fill, and rocks. None of the hearths had stratified deposits.
Pits in structures
Excavators recorded seven general pit features inside structures. These included ash pits,
storage pits, pudding pits and hearths. A brief description of each pit type detected inside
A storage and processing pit was located inside “jacal” 2. As in the case of pit 17, the pit
in jacal 2 was a small (0.40 meters in diam. and 0.27 meters deep) pit which contained a small
ceramic vessel. This long necked pot was buried in terrace fill so as to have the rim of the pot
flush with the occupational surface of jacal 2. Charcoal flecks and ceramic fragments were found
in its interior.
A puddling pit was recorded at the northwestern extreme of pithouse 5. This puddling pit
was given pudding pit number 1. This caliche lined pit has a diameter of 0.46 meters and is 0.10
meters deep. It is possible that this pudding pit served for construction and renovation activities
99
Figure 24. Photo of Hearth 2
100
Table 19. Descriptive Statistics for Hearths/ Datos Descriptivos de fogones
Hearth#/ Loci/ North/ East/ Type/ Tipo Depth/ Length Width/ Plan view/ Fill/ Relleno
Numero Area Norte Este Profun- (diam.)/ Ancho Plano
de Fogon didad Largo
1 E 128.75- 123.60- Unprepared/ 0.15 0.33 0.22 Round/ Ash & ch arcoal/
129.05 124.00 no preparado Circular carbon y ceniza
2 E 130.45- 106.30- Prepared/ 0.22 0.70 0.8 Round/ Ash & ch arcoal/
131.20 107.15 Preparado Circular carbon y ceniza
3 E 127.14- 152.90- Unprepared/ 0.12 0.50 0.45 Oval/ Varied/
127.70 153.60 No Ovalado Varios
preparado
4 E 128.20- 151.35- Prepared/ 0.09 0.64 0.5 Round/ Varied/
128.90 152.00 Preparado Circular Varios
5 E 128.22- 126.80- Unprepared/ 0.15 0.55 0.4 Curved/ Gen. Cult./
129.30 127.55 No Curvado Rell. Cult. Gen.
preparado
6 E 126.00- 122.00- Unprepared/ 0.1 0.70 0.24 Irregular/ Ash & ch arcoal/
126.80 122.70 No Iregular carbon y ceniza
preparado
7 E 121.00- 119.00- Unprepared/ 0.14 1.30 1.1 Oval/ Ash & ch arcoal/
122.50 120.20 No Ovalado carbon y ceniza
preparado
8 E 120.65- 109.15- Unprepared/ 0.12 0.64 0.6 Round/ Gen. Cult./
121.35 109.85 No Circular Rell. Cult. Gen.
preparado
9 E 120.62- ? Unprepared/ 0.11 0.62 0.48 Round/ Ash & ch arcoal/
121.12 No Circular carbon y ceniza
preparado
10 E 123.40- 108.80- Unprepared/ 0.08 0.48 0.47 Round/ Ash & ch arcoal/
123.96 109.35 No Circular carbon y ceniza
preparado
11 E 103.15- 135.00- Unprepared/ 0.13 0.54 0.47 Round/ Ash & ch arcoal/
103.90 135.70 No Circular carbon y ceniza
preparado
12 E 128.20- 86.14- Unprepared/ 0.23 0.78 0.71 Round/ Gen. Cult./
129.06 86.94 No Circular Rell. Cult. Gen.
preparado
13 E 120.00- 122.00- Unprepared/ 0.14 0.43 0.4 Round/ Gen. Cult./
121.00 123.00 No Circular Rell. Cult. Gen.
preparado
14 B4 107.15- 96.15- Unprepared/ 0.12 0.40 0.3 Oval/ Ash & ch arcoal/
107.60 96.62 No Ovalado carbon y ceniza
preparado
15 B6 107.51- 220.71- Unprepared/ 0.3 0.30 0.3 Round/ Ash & ch arcoal/
107.81 221.00 No Circular carbon y ceniza
preparado
101
associated with pithouse 5.
Two were found in pithouses, two in “jacal” structures, and one inside a rock arrangement.
Pithouse 2 included hearth 12 (0.71 x 0.78 meters, 0.23 meters deep) in its southwestern corner.
Hearth 12 was oval. A possible hearth was detected inside pithouse 4. Excavators recorded an
oval depression (0.50 meters x 0.30 meters) with evidence of burning. Since this sub-feature was
not excavated we can only guess at its function. Jacal 4 included in its northern end a small (0.30
meters in diameter) irregular unprepared hearth (hearth 15). Hearth 16 was located inside jacal 5.
Jacal 5's hearth is an oval unprepared depression with evidence of in situ burning. Hearth 16
measures 0.63 meters x 1.00 meters with a maximum depth of 0.80 meters. The crew also
recorded hearth 17 inside rock arrangement 5. Hearth 17 has a diameter of 0.50 meters and is
Middens
Middens are refuse deposits. The crew recorded two middens at Cerro de Trincheras.
Both middens were located in area E. Trincheras middens are visible from the surface as mound
deposited over three small pits. The deposits were stratified with sterile soil between them.
^This indicates multiple dumping episodes. Midden 1 measures 12 m long by 7.3 m wide by
0.52 m deep.
The crew did not excavate this feature in its entirety so the midden’s exact shape could not be
102
Table 20. Distribution of Material by Midden/ Distribución de material por basurero
Midden/ Area Plain Decorated Lithics/ Stone Ground Bone/ Shell/
Basurero Ceramics/ Ceramics/ Litica Tools/ stone/ Hueso Concha
Ceramica Ceramica Herra Litica
Lisa Decorada mientas Pulida
de
Piedra
1 E 5820 7 981 11 3 78 6
2 E 749 0 279 4 5 1 5
that of a petrogylph and is not geometric. This feature appears to have served some special
approximate depth of 0.50 m. All artifact classes we represented in this refuse deposit including
Ramada
A ramada is a small area roofed with a roof of branches, brush and mud that is not walled.
The roof of the ramada is supported with posts. The crew recorded one possible ramada in area
E of the site (Possible Ramada 1 see Figure 26). The ramada was defined on the basis of a post
hole located at N 124.76 E 120.60 and a beam impression burnt into natural caliche. The beam
impression was located directly northeast of the posthole. The post hole measures 0.39 by 0.42
meters and is 0.13 meters deep. Recorders did not find a defined floor or associated sub-features
sub-features. Excavators recovered only a few small lithics and one piece of shell from Ramada
“Petrograbados”
Petrograbados refers to engravings carved deep in large rocks or bedrock (Figure 27).
103
These occur throughout the hill but were not recorded systematically. The carving is deeper than
104
Figure 26. Map of Possible Ramada
105
Figure 27. Photo of Petrograbado
106
purpose, though no function could be confidently ascertained The crew recorded one bedrock
engraving in area B3 (Bedrock Engraving 1 N 109 E 100). The grooves in the bedrock are thick
and linear. The engraving forms a half spiral design. The design measures 2 meters by 1.3
Rock Arrangements
The crew recorded six features as rock arrangements (Figure 28). Rock arrangements
were defined as a single course cluster of rocks arranged in a pattern to a delimited area along a
single level. Rock arrangements are constructed of irregular unshaped local andesite cobbles.
The rocks that define the limits of the rock arrangements tend to be large (greater than 30 cm).
The rock
arrangements can be divided into three different kinds: an alignment of cobbles, linear alignment
of rock clusters, and rock piles. Three rock arrangements were recorded in area E, two in area D,
and one in area B11. Their length ranged from 2.00 meters to 6.00 meters, with a mean length of
3.78 meters. The width of the rock arrangements ranged from 1.5 meters to 6 meters, with an
average width of 3.06 meters. The rock arrangements’ depth ranges from 0.05 to 0.58 meters and
Rock arrangements 1, 2 and 6 are aligned clusters large (greater than 30 cm.) irregular
andesite cobbles. Rock arrangements 1 and 6 are circular while rock arrangement 2 is has an
irregular shape in plan view. Rock arrangement 1 included a caliche lined pit (pit 12) on its
eastern side. The excavation of rock arrangement 2 revealed a portion of adobe plaster from the
entry ramp of pithouse 7. Rock arrangement 6 was constructed against the terrace wall for
terrace 553.
107
Figure 28. Photo of Rock Arrangement 2
108
Table 21. Descriptive Statistics for Rock Arrangements/ Datos Descriptivos de
configuraciones de piedras
Rock Area North/ East/ Shape/ Length Width/ Depth/ Subfeature s/
Arrange- Norte Este Forma / Largo Ancho Profund Subelementos
ment/ idad
Configura-
ción de
Piedras
Rock arrangement 3 was an linear alignment of three rock clusters. The crew found that
an ash filled pit in two of the clusters. The ash pit on the west side of rock arrangement 3 had a
diameter of 0.80 meters and was 0.36 meters deep. The ash filled pit on the eastern end had a
diameter of 0.92 meters and a depth of 0.36 meters. No pit numbers were assigned to these pits.
Rock arrangements 4 and 5 were piles of irregular unshaped andesite cobbles of varied
109
sizes. Rock arrangement 4 was circular, while rock arrangement 5 was roughly rectangular. No
northeastern end of rock arrangement 5. These rock arrangement of piled rock were the largest
of the rock arrangements and probably served a different purpose than the rest of the rock
Burials
The crew identified eleven burial features at Cerro de Trincheras. Two types of burials
were defined: cremations and inhumations. Excavators only located one cremation burial. The
remaining ten burials located at the site were all inhumations. In the section that follows, I
Cremations
The Mapping Project recorded a cremation area in area E. The cremation area was not
included in the research locus. Nevertheless, one cremation located by our excavation in area E
designated as Burial 1( N 99.60- 99.80, E 133.40- 133.60). The excavators described burial 1 as
a small fragment (0.13 by 0.10 meters) of cranium lying upside down under a small
reconstructible vessel. Due to the small size excavators believe this to be the skull of an infant.
The skull fragment was the only bone recovered. The portion of the cremation excavated
measures 0.13 meters by 0.10 meters. Therefore, its is impossible to say what the remainder of
Inhumations
recorded at the site: pit burials, burials in terrace fill and burials covered in rocks from a structure
110
or terrace wall. One inhumations was located in area B2. Another inhumation was found in B4.
Excavators recorded four inhumations in B6. One inhumation was also recorded in each B9,
B11, D, and E.
adult buried in a small oval caliche lined pit (0.55 by 0.44 meters, 0.08 meters deep). The
individual in burial 2 was most probably an infant. The remains were placed flexed in the small
pit and covered with fire cracked rock and charcoal. Most of the fire cracked rocks and the
The crew located Burial 3 (N 104-105, E 89-90) in area B2. Burial 3 was a semi-flexed
semi rectangular pit dug into terrace fill. The body was placed on its back with the knees were
bent to the left side. The left arm was extended. The bottom of the right arm was found resting
on the sternum and neck of the individual. Recorders noted that the individuals lower vertebras
appeared to have been fused together which could indicate some illness. A shell pendant, a green
stone bead, and a plain ceramic fragment were recovered as part of this burial.
The archeologists Burial 4 (N 105, E 115.68-116)in area B4. Recorders describe Burial 4
as a secondary inhumation. The bones were not articulated in a anyway. Only part of the
skeleton was recovered including an arm bone, two ribs, and the cranium. No defined pit was
recorded for this burial feature. The remains extend over an area 0.40 meters squared.
Excavators did not recover any artifacts clearly associated with this feature.
The crew excavated Burial 5 (N 114-115, E 216-217) in area B6. Burial 5 is extended
primary inhumation. The individual was located inside and over the occupational surface of
111
quadrangular stone structure 1. Recorders believe this inhumation most likely post-occupational
in relation to quadrangular stone structure 1. The individual was deposited under a layer of rocks
and soil. Burial 5 runs approximately east-west. No femur or cranium was recovered.
multiple primary inhumation. Archaeologists located burial 6 inside circular stone structure 6.
Two individuals of young age were located next to each other running approximately east-west,
their crania at opposite ends. The individual on the northern side is better preserved than the
other. It had its was laying on its left side with its knees bent. The arms were resting on the
chest of the individual. The other individual was more deteriorated. Only the cranium and the
ribs were clearly identifiable. The head was probably lying towards the west extended on its
right side. Some disturbance was evident to non-anatomical placement of some bones. The
burial of both individual was most probably simultaneous. The excavators noted that burial was
deposited above the occupational surface of circular stone structure 6 and was therefore most
E 237.96-238.08) was located in excavation area B6. The individual was located in the fill of
platform 2. Burial 7 runs north-south with its head toward the southern extreme laying on the
left side in fetal position. The field crew inferred from the size of the bones that this individual
was most likely a sub-adult. Three artifacts were found in association with this burial: a natural
rock which the individual appeared to be holding, and two pieces of groundstone one found
under the lumbar column and the other over the lower abdomen.
112
18 in excavation area B6. Burial 8 was a semi-flexed inhumation in poor condition. This
inhumation runs east west with its cranium towards the east. The individual was laid on its back
with presumably with its knees bent though no leg bones were recovered. Many bones were
disturbed by a rodent burrow which runs through the middle of this burial. Some bones were
also located under the terrace wall that serves as the back wall of circular stone structure 18.
Rocks from the walls of this construction were utilized to cover the body.
under wall fall from the ramp that leads from terrace 277 to terrace 276. Burial 9 is a semi
flexed primary inhumation. This individual was very well preserved. It was laying on its back
with its knees bent towards the left side. All elements were recovered except for the cranium and
the knee bones. Recorders posit the possible absence of these parts due to exposure. The
orientation of burial 9 is east-west with the head to the west. Burial 9 was deposited in a small
terrace fill of area B11. The individual appears to be a sub-adult in a position that was not clearly
defined. The cranium appeared oriented to the west with the frontal side facing south. The lower
extremities appear to have been disturbed after the deposit of this inhumation.
individual. Burial 11 was located directly adjacent to pit 20. The individual was laying on its
back with its knees bent and the arms straight at it sides. The remains were badly disturbed by a
rodent burrow which passed through the middle of the body. Two artifact were recovered in
association with burial 11: a conus ring was found where the right hand would have been and a
113
plain ceramic fragment. The pit in which body was placed was shallow and not well defined
Specialized Features
Three features recorded on the site seem to differentiate them selves from the rest found
at the site. Two of these, “La Cancha” and “El Caracol” are so striking that they have been
described by earlier explorers and researchers who have visited the site (see Chapter 2).
During, the excavation of Cerro de Trincheras, we defined a third feature “El Caracolito” that
appears to have served a different purpose than the rest of the features on the site.
La Cancha
“La Cancha” is located almost at the base of the north side of the hill. This is by far the
most visible feature at the site (Figure 29). This feature is very long and wide in comparison to
walls that serve as its northern and southern extremes. These walls then curve inwards at the
western and eastern ends of the feature. The walls were constructed with dry-laid masonry of
irregular unshaped cobbles measuring up to 2.5 m in thickness and # in height. All evidence
“La Cancha” has a circular stone structure in its interior. This circular stone structure is
located against the middle of the northern wall. It was recorded as a sub-feature of La Cancha
Its walls were constructed in much the same manner as the remainder of the circular stone
structures. This circular stone structure has a diameter of 2.7 meters. The entrance was
identified by a single rock placed lengthwise. The walls at the southern end were quite
deteriorated at the time of excavation. Three apparent occupational levels was defined in the
114
interior of this feature.
115
El Caracol
The “Caracol” (the snail) is the structure that has captivated visitors and researchers the
most (see Chapter 2). “El Caracol” (Figure 29) is one largest structures at Cerro de Trincheras
(12.50 meters by 8 meters, 0.13 meters deep). The Caracol (N 95-104, E 266-280) is located at
the top of the “cerro” in area A1. This structure receives its name from the spiral shape of its
walls (Figure 30) resembling the cross section of a snail shell. The walls of the “Caracol” are
constructed by piling large rocks on the outside and small and medium rocks along the inside of
the walls. The walls are constructed of irregular unshaped cobbles using dry masonry. The walls
stand 1.6 meters on average. Recorders found the entrance to the “Caracol” at its southwest
corner. The single wall begins at the southwest corner, outlines the perimeter and cuts into the
The archaeologists recorded a circular rock arrangement (3.2 by 2.8 meters, 0.5 meters
deep) as a sub-feature of the “Caracol”. The rock arrangement was built against the southern
wall of the “Caracol”. Recorders did not mention finding a floor inside the structure.
El Caracolito
Archaeologists recorded a third specialized feature in area D, “El Caracolito” (the little
snail shell). “El Caracolito” (N 115-121.80, E 68.50- 74.30) measures 7 by 6 meters and is 0.40
meters deep. The walls of the structure resemble a cross section of snail shell in plan view. The
walls were constructed using irregular unshaped cobbles and dry masonry. The spiral effect is
realized by a wall on the interior that runs east- west. Additional to the form, this structure
appears to have served a specialized function , because of the amount of decorated ceramics,
worked ceramics, and shell (n=510) recovered from the interior of this structure. Over one fourth
116
Figure 30. Aerial Photo of La Cancha
117
of the total shell found in structures at the site was found in “El Caracolito”
Summary
This chapter has outlined basic definitions for the 13 feature types and three specialized
features located at Cerro de Trincheras. The description of each feature category includes a
addressed morphology, contents and to some extent function. This architectural classification is
118
Figure 31. Photo of El Caracol
119
CHAPTER 5: RECONSTRUCTING CERRO DE TRINCHERAS
FEATURES
of the site at the time it was built. In this chapter, I tackle this issue two ways. First, I examine
the organization of features on the site. In this section, I conduct a spatial analysis to look at
associations and the distributions of features at the site. Second, I look for possible
consult the ethnographic and ethnohistoric record of Northwest Mexico for structures that appear
The site of Cerro de Trincheras is impressive not only for its size but also because of its
internal organization. The terraces at the site are not identical. The features, including the
terraces, were not built in a haphazard manner. This suggests that the population of Cerro de
Trincheras deliberately organized the space they lived in. The first step in a discussion of site
structure is to consider the spatial distribution of features at a site. This section considers the
The excavators recorded the largest number of features in Area E (n=49). Area E’s
features were also the most varied. Two burials, twenty-eight pits, two middens, three
occupational surfaces, nine pithouses, one ramada, three rock arrangements, and one possible
house floor were recorded in this area. The recorders found the only pithouses in area E. The
pithouses seem to indicate a possible courtyard group similar to those located in Snaketown
(McGregor 1998). The arrangement seems to indicate a common work area (see Wilcox et al.
120
1981 for Hohokam courtyard groupings at Snaketown). The excavators also recorded the largest
number of pit features in this area. Four multi-purpose pits, two caliche lined pits, five ash filled
pits, two pudding pits, thirteen hearths, and two pit features that could not be classified were
found in this area. The number, variety, and nature of features found in Area E further suggests
that it was a major locus of domestic activity for the population of Cerro de Trincheras.
The archaeologists recorded twenty-two features in Area A1. Almost all of the features
in Area A1 are circular stone structures with the exception of one access feature and El Caracol.
Since this excavation area is located at the top of the cerro, it is interesting that almost all the
structures are of the same type. The artifact density of the circular stone structures in A1 is much
lower the circular stone structures in other areas (e.g. the average number of ceramics in circular
stone structures in A1 is 366 while in other areas the average is 2833). Furthermore, the
proximity to El Caracol and the fact that this is the only segment of Cerro de Trincheras that
affords some real privacy from the rest of the site presents the possibility that these features were
Excavation Areas B1, B2, B3, B4, and B7 are very similar in structure. Each of these
terraces has habitation structures, either a jacal or a circular stone structure usually in
combination with pit features. These excavation areas are clustered on the western end halfway
up the north face of the hill. The crew recorded a total of seventeen features total in these areas.
Investigators identified two sets of stairs, connecting one terrace to another, in this section of the
site. Archaeologists located a series of postholes in B7, an ancillary terrace in B3 and a platform
in B4. These features were all open-air structures on which a variety of activities could of been
realized. The inhabitants built one circular stone structure, one platform, and three jacales in this
121
part of the hill. Four different kinds of pits were located on these terraces: a trash-filled pit, a
roasting pit, and two hearths. The high concentration of domestic material and the variety of
associated features suggests that these terraces lent themselves to small scale (in comparison to
many different activities were conducted. Area B6 consists of two long terraces only a few
meters south of La Cancha. The structure of B6 appears to be a larger scale version of B1 thru
B4 and B7. These areas combine habitation structures with open air spaces, on which a variety
of activities were conducted. The crew recorded fourteen features in this area. Four of the
features were burials. Archaeologists identified three circular stone structures, one quadrangular
stone structure, and one jacal. The excavators located a hearth and a storage pit in association
with the jacal. The recorders also identified two platforms and an occupational surface both
McGuire and Villalpando named excavation areas B8, B9, and B10 El Mirador (the
lookout). El Mirador is located high on the north face of the site and includes the terraces with
the highest terrace walls. The crew recorded an ancillary terrace on the back of B8 with a ramp
leading to B9, the terrace above. Excavators located two habitation features on B9, a jacal and a
quadrangular stone structure. A burial was also located on the east side of B9. No features were
located on the uppermost terrace B10. McGuire and Villalpando have suggested that El Mirador
was the “home of the site’s ruling household” (McGuire and Villalpando 1998: 4).
Excavators discovered four features in excavation area B11. B11 is located on the eastern
end of the north face of the cerro. The archaeologists recorded one circular stone structure, one
122
rock arrangement, one pit and one burial in this area. The configuration of B11 is distinct to that
of the habitation areas on the west side of the north face of Cerro de Trincheras. B11 does not
have the same combination of open-air living spaces and habitation structures.
The crew recorded twelve features in Area D. Most of the features in Area D are pits
(n=7). Archaeologists excavated two quadrangular stone structures and El Caracolito in this
segment of the site. Two rock arrangements and a burial were also located in this area. This area
seems to be characterized by the pits that are concentrated in it. One pit was identified as a
roasting pit and another as a hearth. The remaining pits are alike, but I was not able to classify
them.
The site feature distribution indicates five main architectural patterns that may be the
result differential use of space on the hill. Area E represents a complex unit of multiple features
centered around habitation in pithouses and processing in the various pits found in the area. B1,
B2, B3, B4, and B6 can be characterized by a particular residential pattern that combines open air
structures (i.e. platforms, ramadas , and occupational surfaces) with jacales or stone structures.
The small number of features in El Mirador and their configuration suggests that this section
operated as a specialized unit. A1, at the crest of the hill, is characterized by circular stone
structures and El Caracol. The lack of sub-features of a domestic nature seem to indicate distinct
specialized use of the features in this area. Area D is differentiated because of all the pits that
occur in combination with rock arrangements and quadrangular stone structures. The differences
in combinations and densities of features reflect differences the use and organization of these
123
Looking for Superstructures in the Ethnographic and Ethnohistoric Record
The perishable nature of the superstructure of pithouses, jacales, ramadas and stone
structures, at Cerro de Trincheras requires investigators to look beyond the archaeological record
for answers. I consulted ethnohistoric accounts by explorers of Sonora and Sinaloa and
ethnographic reports by anthropologists studying the people of northern Mexico and southern
Arizona. These sources suggest a variety of structure configurations that, when applied to the
The fact that the archaeologists did not recover the superstructure of any of the habitation
features is not surprising. Hot temperatures characterize the Sonoran Desert through most of the
year. Populations of hot dry environments must consider four factors when building: 1)
temperature, 2) humidity 3) rate of air movement, 4) radiation from walls, floors, ceilings and
other surrounding surfaces (Saini 1973). Two basic solutions to the heat exist. One solution is to
build very thick walls which will provide a cool interior by absorbing the sun’s radiation. The
second choice is to build walls which are permeable allowing the cooling wind to sweep through
Trincheras, they chose the latter as evidenced by the impermanence exhibited in habitation
structures.
Mexico are limited. In this analysis I include studies of the Upper Pima (O’Odham and Papago),
Pima Bajo (O’Odham), Opata (Joylraua), Yaqui (Yoreme), Seri (Konkáak), Tepehuan (Odami),
and Huichol. Unfortunately, few scholars provide detailed descriptions of the houses people
lived in. Nevertheless, I believe the accounts provide sufficient information to show what the
124
structures at Cerro de Trincheras might of looked like.
It is not my aim to simply project the ethnographic and ethnohistoric records of Northern
Mexico into the past at Cerro de Trincheras. Instead, I look for structure designs described in
these records which would result in similar material patterns. Ethnoarcheology can be useful
when it “can establish uniformitarian relationships between behavior and its material patterning
that can be taken to the archaeological record with confidence” (Arnold 1998: 8). In the section
that follows I will describe the material patterning for stone structures, jacales, pithouses, and
ramadas recovered at the site and complement these with the ethnographic and ethnohistoric
Stone Structures
Explorers and scholars of Northwest Mexico have described the stone dwellings of
several groups in the region (Spicer 1969; Lumholtz 1904; Grimes and Hinton 1969; Beals 1973
[1932]; among others). Spicer classifies dwellings in the southern highland and central and
southern lowlands of northwest Mexico as “oval mud and stone houses with peaked thatched
roofs” (1969: 788). Other investigators report stone houses for the Huichol, Tepehuan, and
Ópatas. The stone dwellings of these groups share basic construction strategies with those at
Cerro de Trincheras and therefore provide possibilities for the roofing of these structures.
Anthropologists and explorers report the Huichol, Tepehuan, and Ópata had stone houses.
Hrdli ka reports that the Ópatas remember building stone foundations or walls to their
habitations in previous generations (1904: 59). Among the Huichol, Lumholtz describes single
room round stone houses with flat, palm roofs with very low entrances (Lumholtz 1902: 2: 27).
125
Later accounts describe the stone houses of the Huichol with thatched palm or grass roofs
The houses of the Tepehuanes in the 1970s are described as 3 x 4 m stone houses with
posts at the corners to support a brush roof (Figure 32). These are built over a platform
sometimes which measures 10 to 20 meters (Riley 1972:130, 1969: 817). Mason (1971:221),
another investigator, describes Tepehuan houses as constructed of field stone with thatched roofs:
“the roofs are thatched with grass supported on posts , poles and sticks, and the walls generally of
made with dry-laid masonry of irregular unshaped locally acquired andesite cobbles of varying
sizes. The circular rooms measure on average 3.9 m by 3.4 m. The quadrangular stone
structures measure on average 5.34 by 4.44. The walls measure on average 0.60 m high and 0.60
wide at the time of excavation. Some of the wall thickness measured includes some wall
The ethnographic and ethnohistoric accounts suggest a possible configurations for the
roofing of stone structures: thatched flat roofs of grass, palm or brush. Given the lack of post
holes in the stone structures it is most likely that the superstructure was not very substantial.
Lumholtz description of single room round stone houses with flat, palm roofs with very low
entrances (Lumholtz 1902: 2: 27) seems to be consistent with the archaeological evidence at the
site.
126
Jacales and Pithouses
The word jacal in Spanish refers to a hut. Jacales are usually wattle and daub structures
and are very common in northern Mexico. Amsden describes the ubiquitous Sonoran jacal:
“Stones have no place in their construction. Between the corner poles thin shoots, as of ocotillo
or cane are driven into the ground. Withes are interwoven to form an open frame and straw or
some other mat of vegetation is tied onto the frame to compete the wall. Clay is tamped on the
outside against the base of the structure so that water dripping from the roof is carried away
instead of soaking in about the imbedded sticks. Enterprising natives also add a gutter about the
hut” (Sauer and Brand 1931: 115). I have included two other types of dwellings in my discussion
of jacales: brush structures and mat structures. Brush structures have walls that are constructed
of brush(Pennington 1980; Hinton 1969; Spicer 1969; Russell 1972; Hrdli ka 1904). Mat houses
tie mats to poles to use as walls (Pennington 1980: 336, Hrdli ka 1904: 58). These structures are
similar to the wattle and daub structures but do not always include mud or clay as a building
material. The wattle and daub, brush, and mat superstructures would leave a similar pattern in
Pithouses are jacal structures built in depressions. The pithouses at Cerro de Trincheras
conform to the definition of houses in pits, were pit sides are not used as structure walls (Haury
1976: 58). The depth of the depressions can vary greatly. No specific references to houses in
pits were located in the ethnographic and ethnohistoric, yet it is probable that some of the jacal
structures described in the section that follows were built in depressions or the depressions were
127
Figure 32. Southernn Tetephan Stone hose
128
formed by regular wetting and sweeping. The apparent shift to above ground structures
about the time of contact can also be used to explain the absence of these semi-subterranean
structures from the literature of the Sonoran desert. Yet, the construction strategies for the
superstructure of jacales can be used to look for possible designs for the perishable
Investigators commonly report dwellings built of wattle and daub, brush or mats in
Northern Mexico. The Ópata (Sauer 1932: 42; Spicer 1954: 50; Hinton 1969: 881), Pima Bajo
(Hinton 1969:881; Cabeza de Vaca 1944 [?]:63; Pennington 1980: 334), Papago (Hinton
1969:881), Huichol (Grimes and Hinton 1969: 797), and Yaqui (Hrdli ka 1904: 63; Spicer 1954:
12, 1969: 836), built wattle and daub jacales. The Seri (Hinton 1969: 881; Spicer 1969: 790;
Russell 1975: 155; Hrdli ka 1904: 63) used brush structures. Early explorers in Sonora and
scholars describe houses made of mats (Hrdli ka 1904:58) in the region. In the section that
follows I summarize the descriptions of these type of dwellings gathered from the literature of
the region.
The Ópata built wattle and daub jacales (Hinton 1959:17. The Ópata houses were
described during the Diego de Alcaraz party of the Coronado expedition in the mid- 1500s and
later related by Father Mariano Cuevas as 600 mud houses with flat roofs (Sauer 1932: 42).
These were described as “plaited cane walls, sometimes heavily coated with mud, and a roof
composed of layers of cane resting on a few mesquite rafters, with earth piled on top of the cane”
(Sauer 1932:42). Spicer further states that these structures are often accompanied by ramadas
129
that consist of a roof supported on “crotched mesquite posts” (1954: 50). The jacales are less
Anthropologists and explorers also report jacal structures for the Pima Bajo, Papago, and
the Huichol. Hinton describes Onova Pima structures as adobe or wattle and daub dwellings
(1969: 881). Cabeza de Vaca describes the people of the village of Corazones as living in
buhíos(1944: 63). This was interpreted by Pennington to refer to a simple Indian hut (1980:
334). The same kind of dwelling was also described for the Huichol (Grimes and Hinton 1969:
797).
The Yaqui also used wattle and daub dwellings. Spicer describes the Yaqui structures as
dirt covered flat roofs over rectangular cane and mud wattle structures (1969: 836). Hrdli ka
describes the Yaqui jacal as “a fair-sized quadrilateral structure of poles and reeds, or of adobe
reeds or brush, with a flat or more commonly, slightly sloping roof of grass and mud” (Hrdli ka
1904: 63).
The Pima ki (Figures 33 and 34) is also a wattle and daub structure but is constructed in
more substantial way than the other jacales included in this section. Russell does a very
The central supporting framework is usually entirely of cottonwood, though other timber
is sometimes used...The lighter framework is of willow on which is laid arrowwood,
cattail reeds, wheat straw, cornstalks or similar material that supports the outer layer of
earth.
The roof is supported by four crotched posts set in the ground 3 or 4 m apart ,
with two heavy beams in crotches. Lighter cross poles are laid on the last, completing the
central framework. Light willow poles are set half a meter in the ground around the
periphery of the circle, their tops are bent in to lap over the central roof poles, and
horizontal stays are lashed to them with willow bark. The frame is then ready for the
covering of brush or straw. Although earth is heaped upon the roof to a depth of 15 or 20
cm it does not render it entirely waterproof. When finished the ki is very strong and
130
capable of withstanding heavy gales or supporting the weight of the people who may
gather on the roof during festivals (Russell 1975: 154).
These structures are also described near Snaketown. This feature type is taken to provide the
model of construction for the pithouses at Snaketown (Gladwin et al. 1938: 60).
Ségesser in 1737 writes about the Tericopa Pima: “their houses were constructed in the shape of
beehives with small entrances he had to crawl through on all fours when he visited the sick... If
an indian wanted to change the location of his hut, then eight or ten Indians just picked it up and
moved it” (Pennington 1980:336). These structures were constructed by fastening petates
(mats) to a frame of poles that could be moved. Hrdli ka (1904: 58) also makes reference to
early accounts state that “early explorers saw only dwellings made from brush and poles and
palm leaves or mats (petates), and such may be seen among the Sonora natives today”. These
Southern Arizona is the brush house (Figure 35). Among the Seri, Krober describes costal
Sonoran houses as ocotillo frames, tunnel shaped and brush covered (Spicer 1969: 790). Hinton
describes the temporary fishing and gathering camps of the Seri on the coast as “traditional brush
wind breaks”(1969:881). Hrdli ka describes similar structures for the Yaqui: “In the country
districts I have come across an occasional, probably temporary, hut made in the same manner [as
wattle and daub jacales], but entirely of brush and with but a few supporting poles” (Hrdli ka
1904: 63). A brush structure is also described for the Pima Bajo, but these are A-framed
131
structures: “They are used today as summer sleeping quarters and to shield articles from rain.
Two fork shaped poles, each about eight feet long, are put upright in the ground and a ridge pole
is put between them, held in place by the forks. Side poles are affixed to them
132
Figure 33. Photo of Pima Rround House
133
. idgepole about one foot apart to form an A-frame. They are tied to the ridgepole with fiber
cordage or leather thongs. Any kind of straw or brush may be used or the cover... The cover
material is laid on the pole frame and held down by slender saplings anchored transversely to the
of of rock along an elliptical perimeter. The interior of these structures was cleared of large
rocks. The material remains of these structures (Figure 36) are consistent with the proto-historic
Pima sites recorded in the Santa Cruz, San Pedro, and Sonoita Creek valleys of southern
Arizona: “the locations where structures once existed are delineated today by elongated stone
rings that measure approximately 6 by 12 feet. Rock sizes of 4 to 6 inches are typical and
spacing between rocks is consistent in the structures of all three valleys” (Seymour 1993: 3).
These ephemeral structures could have been built in almost all of the configurations I
present from the ethnographic and ethnohistoric record. The description of the Pima ki seems to
be the least likely configuration for the jacal structures at Cerro de Trincheras. A structure able
to maintain the weight of several people during festivals would most likely require quite a bit of
soil as part of its construction material and substantial roof support. Since the jacales at Cerro de
Trincheras do not have that much fill it is more likely that they took another less substantial
form. Also the Pima ki is built with many thick posts along its perimeter. These posts would
most likely leave a pattern in the archaeological record. The excavators found no such post hole
134
Figure 35. Arrow Bush Kitchen Area and Pima
Woman
135
Figure 36. Elliptical Rock Arrangement ojn
early Piman Sites
136
Most wattle and daub structures are similar to each other differing only by building
materials. Any of these wattle and daub dwellings could have left a pattern such as the one
archaeologists found for jacales at the site. The same can also be said for the brush and mat
walled structures. Ségesser description of mat walled structures as moveable (Pennington 1980:
336) presents certain possibilities for the interpretation of jacal features at Cerro de Trincheras.
Movable structures would leave subtle traces in the archaeological record such as is the case at
Cerro de Trincheras.
Pithouses at Cerro de Trincheras are also quite difficult to difine. The excavators of
pithouses at the site defined pithouses by the presence of one or more of the following: a slight
depression which averaged 0.38 m dug into native caliche; caliche manipulated to create a flat
occupational surface; and ramp entries. The pithouses measured on average 3.46 m by 3.21. No
postholes were located in association with these features. The archaeologists located medium
sized rocks above the pithouse use surface. These rock were probably used to weigh down
roofing material. As with the jacales, the absence of postholes seems to suggest a less substantial
building method than the Pima ki. The superstructure could of been built of brush, mats, or
wattle and daub in any of the configurations present in the ethnographic and ethnohistoric
literature.
Ramadas
the Sonoran desert. These structures provide open-air shelter from the sun. Ramadas are roofs
137
Ramadas in the ethnographic and ethnohistoric record
According to the literature of the region, the Pima Bajo built ramadas. The Pima Bajo’s
ramadas are constructed by “[f]our posts usually with forked tops... placed upright on the ground.
Beams are anchored to the post tops with leather thongs or with vines, and brush is added to a
frame of saplings arranged across the beams” (Pennington 1980: 337). Russell describes the use
“the roof furnishes a convenient place for drying squashes, melons, fruit, and in the old days
cotton, where the dogs and poultry can not disturb them. Under its shade the olla of drinking
water is set in a crotched post or is suspended from above by a maguey fiber net. Here two
parallel ropes may be hung and a cloth folded back and forth upon itself across them, thus
forming an impromptu hammock in which to swing the baby. Here the metate and mortar are
usually seen, and here the women sit and weave baskets or preform such other labors as may be
done at home. It is the living room throughout the year around, and now that the fear of Apaches
The ramada is rather common in the Sonoran desert. Nevertheless, the excavators
recorded only two possible ramadas. The archaeologists defined the possible ramada on the basis
of a post-holes and a beam impression burnt into caliche. It is possible that more ramadas were
present at the site and let no observable trace of their presence. We base this on an experiment
conducted by the crew that sought to build a ramada without digging the posts into the ground.
The crew built the ramada by making a roof of ocotillo and palm and fastening this roof to poles
138
Figure 37. Huichol Ramada
139
of palo verde. The poles were supported by large cobbles from the hill. All construction
materials were locally acquired. The ramada stood up quite well until it was eaten by cows
for the features the crew excavated at Cerro de Trincheras. The stone structures were most
probably covered by a flat thatched roof. The roof could have been constructed of a variety of
lightweight material including, grass and brush. The jacales and pithouses were most likely
140
CHAPTER 6:
ARCHITECTURE
have made interpretations of Trincheras without basic baseline research. Investigators involved
in the Cerro de Trincheras Research Project have uncovered enough data to address basic
questions. The data recovered from the excavation of the site make a more thorough comparison
between the Hohokam and Trincheras possible. In this chapter, I apply the data of the
architectural features at Trincheras and compare these to the archaeological types previously
Hohokam Overview
Hohokam reside in the Sonoran desert in southern Arizona and Northern Sonora. Generally, the
Hohokam built their villages along major rivers in the region. Investigators have interpreted the
Phoenix Basin, along the Salt and Gila Rivers, as the “core” area for the Hohokam (McGuire
1992: 4). Scholars have understood the remainder of Hohokam area as “peripheral.” Hohokam
researchers include the Tucson Basin and the Papaguería as Hohokam peripheries. Archeologists
have noted that each of the Hohokam peripheries has its own developmental sequence and
material pattern yet, participated in variable ways in the Hohokam system (Doelle et al. 1987;
141
Doyel 1976; Fish and Fish 1977).
The early part of the twentieth century saw the birth of research on the Hohokam.
Investigators derived the chronological sequence for the Hohokam primarily from excavations at
the site of Snaketown (Gladwin et al. 1938). Since then, many investigators have revised the
Hohokam chronology (Bullard 1962; Plog 1980; and Schiffer 1987; among others). Though the
problems in Hohokam chronology are far from resolved, Dean (1994:91) has championed a
chronology based on the radiocarbon and archaeomagnetic dates from by large scale contract
archaeology projects, and the tree-ring sequence recently established for the Tucson Basin. Table
? shows Dean’s chronology and several phase sequences for the Hohokam.
Archaeologists call the earliest Hohokam period the Pioneer period (ca. 300 BC to 700 AD). The
characterize the Pioneer period (Doyel 1987: 7). During the Colonial period (ca. AD 700 to 800)
the Hohokam moved into peripheral areas, and sites became larger and more common. The
Colonial period Hohokam centered many of their sites around communal architectural structures
(i.e., ball courts and platform mounds). During this period, the Hohokam builders expanded,
narrowed and deepened irrigation canals. Cremation was the preferred method of burial. The
Sedentary period (ca. AD 800 to 1100) sees a sharp change in ceremonial architecture, ornate
craft production industries develop. Although cremation burial is the most prevalent, extended
Researchers have documented major changes for the Hohokam during the last Hohokam
period: the Classic. The period saw the expansion of irrigation systems, a change from houses in
142
Table 22. Hohokam Chronologies (adapted from Cordell 1997:200 and Haury
1976:39)
Date Tucson Phoenix Santa Gila Bas in Papagu ería
Period Basin Basin Cruz (Dean and (Dean and
(Dean (Dean (Downum Greenleaf Greenleaf
1994:91) 1994:91) 1993:23) 1975:12) 1975:12)
1500 Proto-
historic
1450
Civano Unknown Civano
1400 Tucson Hohokam Sells
1350
Classic
1300 Soho
Tanque
1250 Verde Early Soho
Classic
1200 ?
1150 Topawa
Sacaton
1100 Rincon Sedentary
1050
Sedentary Sacaton
1000 ? Santa Cruz Late
Colonial
950
Gila Butte
900 Vamori
Early
850 Colonial
Rillito
800 Snaketown Santa Cruz
750 Late
Colonial Pioneer
700
Sweetwater
650
600 Estrella
500
450
Vahki
400 Pioneer Snaketown
350
300
143
pits to above ground structures, changes in ceramic styles, and a shift to inhumation as the
preferred burial method (Doyel 1987: 7). Classic period Hohokam was contemporaneous with
Past two decades have seen a real explosion in Hohokam research. Cultural resource
management firms have conducted most recent work due to the rapid development of Southern
Arizona. The massive data has allowed investigators to document incredible variability within
the “Hohokam” world. Despite the influx of new information, research conducted at Snaketown
still informs much archaeological work today (Doyel 1987: 10). Hohokam archaeologists still
use the basic architectural categories Haury established at Snaketown (Martynec 1993; Mitchell
Sayles (1938) established the first Hohokam architectural typologies following the
excavation of Snaketown. Sayles identified a trend in Hohokam architecture from large square
and elliptical dwellings with formal sub-features (e.g., step entries). Later Haury (1976:44-77)
refined Sayles’ house typology based on the excavations at Snaketown in the 1960s. Haury
created the architectural typology employed throughout most of the Hohokam area today (1976:
45-77). In the section that follows, I summarize the residential architectural trends reflected in
Haury’s typologies and compare these types to the habitation architectural types present at Cerro
de Trincheras.
The Pioneer period witnessed a wide range of architectural variability. Haury defined -4
144
Figure 38. Emil Haury’s Housing Sequence for the Hohokam
145
seven house types for this period based on Snaketown (1976: 68). The Hohokam built P houses
during the earliest part of the Pioneer period are square with rounded corners and large (10 to 15
m in diameter). The builders constructed houses in pits by digging a shallow pit to caliche.
Floors were prepared with a mixture of caliche and clay about 3 cm thick. Floor area for P-4
houses averaged 51 m2. These houses had two sloping entries. Four large posts supported the
roof (post holes were 1 meter deep) as well a several smaller interior posts. The pre-Hispanic
inhabitants constructed the walls with 10 to 15 cm in diameter posts every 20 to 50 cm. P-4
In the remainder of the Pioneer period the Hohokam built six other kinds of houses and a
smaller version of the P-4 house. The P-3 were square in plan, built on native soil, a slight floor
edge groove with post for walls along the groove, when present. Main four posts supported the
roof set almost at the corners of the structure. Entries were parallel sided with no step. The floor
area averaged 22 m2. The Hohokam constructed P-2 houses in the same manner as the P-3
houses but were rectangular in plan view. The roof of P-2 houses was supported by a roof of
posts one meter from the edge of the structure. P-1 houses were rectangular in plan view with
rounded edges. Floors were prepared with a mixture of caliche and red clay. P-1 houses had one
parallel sided entry. The Hohokam constructed the walls with mesquite posts 10 cm in diameter
set 25 cm apart which the builders later covered with brush and a layer of clay. They generally
aligned the interior roof supports in three rows parallel to the long axis of the structure. A hearth
was found midway between the center of the structure and the edge. The floor area averaged 11
m2 .
Haury defined two smaller house types for the Pioneer period. P-6 houses were small
146
(with a floor area of 6.2 m2) rectangular structures with a well-prepared floor and no hearth. The
P-6 structures had a pronounced floor groove to”engage the butt ends of inner reed lining”
(Haury 1976: 68) and outside posts. P-5 structures were the same as P-6 structure without a
hearth and with an end entry instead of a side entry. The P-5 structures had an average floor area
of 8 m2.
P-7 houses are the only true pit houses used by the Hohokam. The P-7 house uses the
edges of the pit as side walls for the structure. A p-7 house is square with rounded corners and a
prepared floor of a caliche and clay mixture. Pit sides rose 0.05 meters high and were rounded
toward the top. Several roof supports were placed inside the house but no wall post holes were
found along its perimeter. The P-7 house had a side entry with no step. The floor area for these
The Colonial period was a period marked by Hohokam expansion. The period did not
witness much change in domestic architecture. The Hohokam builders made only slight
modifications to pervious architectural styles. Haury defined three house types for the Colonial
period (1976: 65). The C-1 house is similar to a P-1 house with a well defined post hole pattern
along its perimeter, no internal roof supports and no groove along its edge. The C-2 houses are
rectangular with a floor of a caliche and clay mixture and an average floor area of 20 m2. The C-
2 house type has a parallel sided entrance near the mid point of the longer wall. Archaeologists
often found steps at the entrance. C-2 entrances outside Snaketown have been found with stone
risers at entrances. Internal posts supported the roof parallel to near the edges of the house. The
Hohokam constructed the walls with a line of mesquite posts along the perimeter 10 cm in
147
diameter and 25 to 50 cm apart, filled with reeds and brush and capped with a layer of clay. A
well-made hearth is normally associated with these structures near the entry. The C-3 house type
is rectangular but shorter than the C-2 house. C-2 house floors were either prepared with a
mixture of soil or built on native soil. Hearths were prepared clay-lined basins near the entry. C-
3 houses are smaller and more formalized versions of the larger Pioneer period house types.
architecture during the Sedentary period. Ball courts and platform mounds become more
common. Nevertheless, changes in domestic architecture are slight. Haury defined five house
types for the Sedentary period. S-1 houses are elongated with rounded corners. The floors are
prepared with a clay-caliche mixture with an average thickness of 0.05 to 0.15 m Floor edges
sometimes had small grooves or lips. The average floor area for S-1 structures is 20 to 25 m2.
Entries were positioned halfway along the long side of the structure. These were typically
bulbous shaped. The Hohokam sometimes built steps several different forms: a step of a caliche
flooring mixture with a molded sill at the top, with a slab of mica schist or of risers made by
stacking small logs. Interior roof support was highly variable for this house type. No
overwhelming pattern was evident. In a few cases no evidence for internal roof support was
found. The builders constructed side walls in the same manner as C-1 houses with a series of
mesquite posts along the perimeter. They fastened a reed and brush covering onto the posts, and
covered it with clay. Clay-lined hearths with an average depth of 0.25 m2 were near the entry.
Trivets used in cooking were often found in S-1 structures. The S-2 house is larger, square with
rounded corners and marginally convex sides. The floor area for S-2 houses averages 42 m2.
148
Entries to S-2 dwellings are typically sort and parallel sided. The roof was flat, based on the four
major posts with auxiliary posts near the edge of the floor to support the side wall posts. The
Hohokam built side walls as in S-1 structures. The builders always placed hearths near the entry.
Haury defined S-3 Haury as big versions of S-1 structures. The floor areas for S-3 houses
average 52 m2. Due to large size three rows of roof supporting posts were placed in the interior.
Sedentary period Hohokam built parallel sided entries with occasional stone risers.
Haury defined the S-4 type as a small, elliptical structure built in a pit 0.06 m below
ground surface. The floors were prepared with a mixture of clay and caliche and had an average
area of 8 m2. The Hohokam sometimes built entrances with a large step. Clay lined hearths were
placed near entries. Three aligned postholes were found in S-4 structures. The Hohokam built
the side walls of a caliche and clay mixture puddled 0.15 m thick and rising to a maximum height
of 0.65 m.
Haury recognized gave the designation of S-5 to structures that were rectangular with
square corners and adobe walls 10 cm thick. At the time of excavation only 0.10 m of the walls
was preserved. Haury inferred that these walls sloped upwards to “support horizontal stringers”
(1976: 62). The S-5 structures had an average floor area of 7 m2. Hearths varied from burned
areas on the floor to formal clay-lined hearths. Investigators have interpreted this house type as
the source for the construction methods later employed in domestic architecture during the
The Hohokam changed many aspects of their way of life during the Classic period. The
Hohokam area saw a major territorial reduction during the Classic period(Gummerman and
149
Haury 1979: 86). Major changes in domestic architecture echoed the vast changes in the Classic
period. The classic period Hohokam employed new methods of walls construction that included
solid clay walls and post reinforced adobe walls (Gummerman and Haury 1979:86). Though the
Hohokam continued to build houses in pits, the clay walled and post reinforced structures built
above ground became more common. The above ground structures were often contiguous rooms
and in the late part of the phase develop into multi-storied structures.
The Hohokam abandoned Snaketown before the Classic period, so Haury looked at a
nearby Classic period site to elaborate his architectural typology for the period (1976). Classic
period houses were built on desert surface or below with house floors of prepared clay. Classic
period structures were rectangular with squared corners and an average floor area of 15 m2.
Walls vertical, built of puddled adobe with an average thickness of 0.25 m. Typically, the
builders internally reinforced the walls with a series of regularly spaced posts 10 to 15 cm in
diameter. Entries did not include passages as in earlier Hohokam house forms. Internal roof
supports were variable or not detected. Hearths were centrally placed and clay lined. Haury
defined two house types unique to the period Cl-1 and Cl-2 (1976). Cl-1 houses stood alone
while Cl-2 houses were contiguous sets of rooms. Compound walls also characterize this period.
The Hohokam often built Classic period structures on top of clay capped mounds that stood two
meters or more in height (Gummerman and Haury 1979: 87). The multi-storied structures or
“great houses” become more common toward the latter part of the Classic period. The great
Hohokam building strategies change through space as well as through time. The regions
150
outside the Hohokam core exhibit quite a bit of variation. Yet, one factor is unmistakable,
though architectural styles might vary, they still share a basic Hohokam pattern. For that reason,
architectural typologies that were created for the Phoenix basin are still applied at Hohokam sites
outside the core. In this section, I touch on the architecture of the Tucson Basin and Papaguería
The Tucson Basin refers to the region surrounding the Santa Cruz River in Arizona. The
Tucson Basin extends from the Tortolita to the Santa Rita Mountains. The Hohokam appear in
this area during the Sweetwater phase (see Table 21) And their population in this area climaxes
during the Rillito phase (Doelle 1985: 15). Generally, Hohokam architecture looks much like the
architecture of the Phoenix Basin. Though variation is visible when it comes to shape and size,
Tucson Basin domestic structures are similar to those in the Phoenix Basin. Even structures built
Structures excavated at Linda Vista Hill, Fortified Hill and Cerro Prieto, cerro de
trincheras sites look much more like traditional Hohokam houses than those we have excavated
at Cerro de Trincheras. At all three sites investigators excavated structures that from the surface
resemble Cerro de Trincheras’ stone structure. The archaeologists after excavation described
these structures as “masonry outlined pithouses” (Downum 1993: 67; S. Fish et al. 1984;
Downum 1986, Greenleaf 1975). Downum describes Feature 4 at Cerro Prieto as a dwelling
with a prepared floor, an entryway, and numerous domestic artifacts. In structures at all three
sites excavators report post holes for roofing support in the interior of these structures. These
The Papaguería is the lesser know branch of the Hohokam. Papaguerían Hohokam lived
151
in an area bounded by the Gila Valley to the north, the Santa Cruz River Valley to the east, the
Colorado River valley in the west, and Caborca, Sonora, Mexico to the South. The Papaguería is
devoid of permanent streams. The material pattern of the Hohokam in the Papaguería led Haury
(1976) to develop his Desert Branch of the Hohokam. Very few excavations have been
conducted in the Papaguería. Archaeologists that have excavated in this area have uncovered
architectural patterns though not identical to the Phoenix Basin, still resemble it in many ways
The Hohokam of the Papaguería built pithouses that were generally rectangular in plan
view with rounded corners (Withers 1973; Rosenthal et al. 1978; Marmaduke and Martynec
1993). Though forms and post hole placement was less formalized, they still fall within general
pattern for the Hohokam. Floor of Papaguerian houses were generally plastered. Hearth
placement, as in other Hohokam house was typically near the door. Entries were either parallel
sided or bulbous. A few informal feature types have been observed at Papaguerian sites
(Martynec 1993: 119). Though Martynec refers to these structures as “ephemeral”, they had
prepared floors, and in some cases hearths and posts for roofing. These structures still appear
The Hohokam had a variety architectural styles that changed greatly through time.
Despite the variation, some serious differences are apparent when we compare the domestic
architecture of the Hohokam with that of Cerro de Trincheras. The contrast results from
differences in basic construction strategies. Hohokam builders constructed more formalized and
substantial structures.
152
The Hohokam builders built houses in several forms, yet, they all share some basic
construction elements which are not present at Cerro de Trincheras. They typically prepared a
floor with a mixture of clay and caliche. No preparation of this sort was used for any of the
habitation structures at Cerro de Trincheras. The lack of preparation made floors difficult to
identify. Though several patterns for roof supports were identified for the Hohokam, at Cerro de
Trincheras no interior roof supports were located in pithouses, jacales, or stone structures.
Hearths were identified in almost all Hohokam structure types. Generally, Hohokam hearths
were prepared with a well fired clay lining. Hearths are rare in Trincheras structures; of the 65
jacales, pithouses and structures only 13 percent of these had hearths. All the hearths located in
the interior of Trincheras residential structures were unprepared. No wall posts were identified
along the perimeter of any of the jacales or pithouses at Cerro de Trincheras. Almost all
Hohokam house types had a pattern of post holes along their perimeter. The entries of the
pithouses at Cerro de Trincheras are usually indicated by ramps. Entries into Hohokam
dwellings vary in form and construction material. Floor area of Hohokam domestic structures is
highly variable. Table ? summarizes average floor area for Hohokam house types and Cerro de
Trincheras structures. The typical Hohokam dwelling house was more formally prepared and a
The large body of information generated by the growth of cultural resource management
has allowed Hohokam researchers to revisit long standing debates of Hohokam origins. New
data has strengthened arguments that see the Hohokam as an indigenous development with
antecedents in the San Pedro Cochise (Wilcox and Shenk 1977; Wilcox 1979). San Pedro
153
Table 23. Average Floor Area for Hohokam and Trincheras Domestic Structures
Hohokam House Type Average Floor Area (Haury Cerro de Trincheras Structure Average Floor Area
1976) Type
P-6 6.2 m2
S-5 7 m2
S-4 8 m2
P-1 11 m2
C-1 11 m2 Pithouse 11 m2
Cl-2 15 m2
C-2 20 m2 Jacal 18 m2
S-1 23 m2
P-2 22 m2
P-3 22 m2
S-2 42 m2
P-4 51 m2
S-3 52 m2
C-3 unknown
Cochise origins have also been assumed for the Trincheras people. In this section I examine
Hohokam researchers addressing this question have grappled with the apparent
dissimilarity between the earliest Hohokam architecture and the San Pedro Cochise (Cable and
Doyel 1987: 57). Differences in architecture surround the formal nature, and large size of houses
built in the Vahki phase. Recent contract work has allowed Hohokam researchers to postulate an
earlier phase for the Hohokam: the Red Mountain phase of the Pioneer period in the Phoenix
Basin and Cienega phase in the Tucson Basin . During the Red Mountain period architecture
can be described as small square houses with rounded corners (Cable and Doyel 1987: 59). Red
154
Mountain houses have wall trenches, plastered floors and no evident pattern for roof supports.
The Red Mountain phase allows Hohokam researchers to bridge the material gap between the
San Pedro Cochise and the Pioneer period Hohokam. Researchers have also looked to the
Cienega phase (ca. 800 BC- AD 150) found along the Santa Cruz floodplain to remedy the
material gap leading to the Hohokam (Mabry 1997: 1). Gregory (1997:2 ) has described the
architecture for the Cienega phase as circular pit structures, three to four meters in diameter, with
a series of evenly spaced post holes along their perimeter. Floors were not prepared and no
In contrast to differences between the San Pedro Cochise and the Hohokam, Trincheras
architecture shares certain similarities with San Pedro Cochise. The houses of the San Pedro
Cochise have been described as oval or circular structures with informal hearth areas and no
evidence of structural support (Sayles 1945: 3-4, 1983: 125-129; Cable and Doyel 1987: 58).
Cable and Doyel have characterized these dwellings as giving the “overall impression ... of
impermanence and only seasonal habitation” (Cable and Doyel 1987: 58). These structures
sound very much like the jacal and pithouse structures excavated at Cerro de Trincheras.
Trincheras research has traditionally been undertaken under the shadow of a Hohokam
outsider of the Hohokam system have been undertaken in the absence of fundamental baseline
research. The Cerro de Trincheras Research Project has produced the kind of data necessary to
Hohokam researchers have revised the chronology several times. Recent data produced
155
by large contract archaeology project has allowed the chronology to be refined on the basis of
chronometric dates and a tree-ring sequence for the Tucson Basin. The chronology for the
Hohokam establishes four periods: Pioneer, Colonial, Sedentary and Classic. Much variability
has been documented within the Hohokam world, especially in the Tucson basin and the
cannot deny an overarching Hohokam material pattern still exists. The research conducted at
Snaketown still informs much of Hohokam research. The architectural classification mostly
based on early Snaketown research is still widely applied in the Hohokam area.
different house types built between the Pioneer period and the Classic period. The Hohokam
houses typically have plastered floors, in the pre-classic walls are generally built of posts covered
with brush and reeds covered by a clay layer, in the Classic period walls were constructed of
puddled adobe walls, formal hearths were placed near entries during the pre-Classic and
centrally located during the Classic period, in most Hohokam house types roofs were supported
by a pattern of large interior posts resulting in either flat or gabled roofs, entries were parallel
sided or bulbous in the pre-classic while no passages were used during the Classic period. Sizes,
shapes and configurations of entries of Hohokam house types were highly variable.
Generally, Hohokam houses are more formal and more substantial than domestic
architecture at Cerro de Trincheras. The floors of Cerro de Trincheras were very difficult to
detect in most cases while Hohokam house floor were typically prepared. The sidewalls of both
pithouses and jacal structures at Trincheras show no evidence for posts, puddled adobe walls, or
post reinforced adobe walls set along the perimeter while almost all Hohokam house types
156
exhibit evidence of a substantial wall construction. Post supports for roofing were usually found
in the majority of Hohokam house types, no internal roof supports were found in stone structures,
jacales and pithouses at Cerro de Trincheras. Hearths in Hohokam dwellings are generally
prepared with a clay lining and placed near the entries, while Trincheras hearths in structures are
rare, very informal and unprepared. The typical Trincheras dwelling is much more ephemeral
The ephemeral nature of Trincheras domestic architecture is similar to that of the San
Pedro Cochise and the proto historic Pima. In both cases houses are typically, round to oval in
plan with informal hearths an no evidence for structural support. Additionally, both housing
strategies leave the impression of evanescence. Both the San Pedro Cochise and the Early Pima
settlements have been described as “rancheria” style. These settlements are dispersed and
involve little planning. This is certainly not the case at Cerro de Trincheras. The sites internal
organization is very complex. Further research needs to done in order sort out the architectural
157
CHAPTER 7: SUMMARY AND CONCLUSIONS
This thesis represents the first feature typology for Cerro de Trincheras. A study of this
kind provides basic baseline information about the site and the little known Trincheras tradition.
It represents a crucial step toward understanding the organization and use of space at Cerro de
Trincheras. Furthermore this study provides feature categories that will lend themselves to the
comparison within the site, across the Trincheras area, and beyond.
Our team excavated one-hundred and twenty-four features at the site of Cerro de
Trincheras during the 1995 and 1996 field seasons. The excavation permitted the identification
of 14 feature types and three specialized features: access features, burials, circular stone
pithouses, platforms, a possible house floor, a possible ramada, rock arrangements, and pits. I
made the feature classification primarily based on morphology and building methods.
I identified three feature types, pithouses, jacales, and stone structures, as habitational
structures. Stone structures are the most numerous features at the site after the terraces. These
features have dry-laid masonry walls of andesite cobbles. They were circular or quadrangular in
plan view. Stone walls provided a base for a perishable superstructure. The crew excavated
twenty-one circular stone structures and four quadrangular stone structures. The excavators
Trincheras pithouses are house-in-pits. The crew excavated nine pithouses at Cerro de
158
Trincheras. The inhabitants built these features in shallow depressions cut into native caliche.
Floors of the pithouses are unprepared and generally difficult to identify. Ramp entries provide
access into seven Trincheras pithouses. Two pithouses had hearths and a third had a pudding pit
inside.
The crew excavated eight jacal structures at the site. Trincheras jacales appear as an
support for a perishable superstructure. The lack of a contiguous post support pattern along the
perimeter suggests they constructed the side walls of a light material such as ocotillo, brush or
mats. The use surfaces of jacales were unprepared and were generally identifiable as
discontinuous changes in texture or a layer of small cobbles covered by soil to create a level
surface. One hearth and two storage and processing pits were recorded in association with these
structures. The hearth was small, irregular, and unprepared. The storage and processing pits
were identified as pits with jars buried under the use surface of the structure.
The archaeologists identified a feature as a possible house floor. This feature did not
provide enough information to assign it to any of the previously discussed residential structures
confidently. This feature is an area 8 m2, irregular in shape, of burnt caliche. The surface was
0.15 m thick and is very red in color. The reddish coloring implies that the Trinchereños fired
I have classified two feature types as open air work areas: ramadas, and platforms. The
crew identified one possible ramada. The crew identified this feature by post holes and a beam
impression burnt into native caliche. Platforms are flat surfaces built toward the fronts of
terraces. Platforms were delimited by a row of rocks instead of a rock wall like a terrace. High
159
concentrations of artifacts were found on the surfaces of these features.
The feature class Occupational Surface was defined as a hard packed surface or a
concentration of artifacts along a level. The crew excavated four occupational surfaces at the
site. The archaeologists described two occupational surfaces as concentrations of artifacts along
a level. They identified two other occupational surfaces as hard packed surfaces. The pre-
arrangements ranged in from, rock piles, to circular arrangements, to scatters of surface rocks.
The investigators excavated six rock arrangements throughout the site. Two of these structures
had pits associated with them and a third had a hearth at its northeast end. I could not
The team recorded a total of 43 pits at Cerro de Trincheras. I divided the category “pit”
into three general categories: pits, puddling pits and hearths. The crew excavated twenty-four
pits, two puddling pits, and 17 hearths at Cerro de Trincheras. I further divided the general
category pit into six different categories: multipurpose pits, caliche lined pits, ash filled pits,
roasting pits, trash filled pits and storage and processing pits. The classification the pits types
were based primarily on morphology and pit contents. The crew recorded four multipurpose pits,
two caliche lined pits, six ash filled pits, two roasting pits, one trash filled pit, and one storage
Access features form an essential part Cerro de Trincheras. The project permitted the
identification of two types of access features: stairs and ramps. Access features lead from one
terrace to another. The excavators recorded two sets of stairs and two ramps during the
160
excavation of the site. Ramps were constructed as sloping surfaces by building a wall and filling
it with a rubble core that they then covered with a thin layer of soil. Trincheras people built stairs
as rock alignments with fill behind them to rise as steps toward a higher level.
I assigned the feature type petrograbado to deep carvings into bedrock or large rocks on
the hill. These “carvings” are deeper than petroglyphs and do not form geometric designs.
Though these features are found throughout the hill, the crew recorded only one of these. No
The recorders identified two middens at the site. The middens are mound shaped
deposits of refuse material. Both of the middens were in area E. One of these features had
stratified deposits which suggest multiple dumping episodes. This midden was deposited over
three former pits. We did not excavate the other in its entirety so its exact shape was not
determined.
The investigators also recorded burials at Cerro de Trincheras. Both cremations and
inhumations were represented at the site. I further classified the inhumations as pit burials,
burials in terrace fill, and burials covered in rocks from a structure or terrace wall. The crew
A spatial analysis of the feature types and their distribution reveals several patterns for the
site. A residential area is located on the middle to western side of the northern face of the terrace
(excavation areas B2, B3, B4, B6, and B7). These areas are characterized architectonically by
the presence of jacales and circular stone structures in combination with open air work areas
such as platforms. My observations of the architectural patterning are also consistent with
O’Donovan’s (1997: 209) depiction of the northwestern slope of the hill as the main residential
161
area.
The crest of the hill has a distinct pattern that marked by many circular stone structures,
the presence of El Caracol, and a low density of artifactual material. This pattern in combination
with its private location seems to suggest a ritual or ceremonial use of the features in this area.
Further support can be gained from O’Donovan’s (1997: 197) spatial analysis of the organization
and construction of terraces at the crest and found that these were used “to augment natural
features, to create certain images, enclose areas and provide vantage points.”
The size, location and low artifact density suggest that La Cancha represents its own
architectonic unit. This feature served a very public ritual purpose. La Cancha is a unique
architectural structure that is visible from any terrace on the northern side of Cerro de Trincheras.
O’Donovan (1997: 203) suggests La Cancha served as an arena for public performances of some
type.
Another feature pattern is present in El Mirador. The three terraces that make up El
Mirador are connected and seem to represent a unit. The lowest terrace has a ramp and an
ancillary terrace along the back. The next terrace has two residential structures, a jacal and a
quadrangular stone structure, and the final terrace has no features on it. This pattern is unique to
this area and probably represents a specialized use. McGuire and Villalpando (1998: 4) suggest
Excavation area B11 shows another architectural pattern. The excavators recorded a rock
arrangement, a circular stone structure, and a pit in this area. This arrangement is distinct to that
of the northwestern slopes, yet show evidence of residential occupation. This area possibly
served as a residential area with lower occupational density. O’Donovan (1997:210) further
162
supports this explanation by indicating that the eastern slopes are characterized by a low artifact
density.
Area E exhibits its own complex pattern of residential use. The crew recorded the largest
number to features in this area and the widest variation in feature types. Pithouses were only
found in this area. The combination of pithouses and a variety of pits proves that this area was
The excavation resulted in the identification of area D as its own architectural unit. This
area includes several narrow terraces used for agricultural purposes, pits, rock arrangements,
quadrangular stone structures and El Caracolito. The pits and the quadrangular stone structures
suggest low intensity residential use, in combination with specialized activities associated with
I consulted the ethnographic and ethnohistoric record for possible types of superstructures
for the jacales, pithouse, stone structures and ramadas at Cerro de Trincheras. In all the above
cases, the lack of formalized roof and wall supports suggest that these structures were walls and
roofed with light materials. In Sonora and southern Arizona ethnographers and explorers report
structures walled with light materials. Building materials, range from brush, to reeds, to mats.
The material was then fastened onto poles and often covered with a layer of mud. This type of
structure is ideally suited for hot dry environments because it maximizes the rate of air flow
They most likely built the jacales, pithouses and ramadas at Cerro de Trincheras in one of
these ways. The poles were most likely built of ocotillo because of the low quantity of trees in
the valley. Brush, mats or sticks were probably attached to the poles to form walls. A mud
163
covering may or may not have been applied. The people of Trincheras probably roofed the stone
In this study I compared the traditional Hohokam architectural types to those I defined for
Cerro de Trincheras. The comparison discovered major architectural differences between the
Hohokam and Trincheras features. Hohokam features varied both through time and through
space. Yet, all of them share basic Hohokam building characteristics. No Trincheras structures
resemble the above ground adobe structures of the Classic period so I focus on the pre-Classic for
this summary. Hohokam structures have well prepared often fired floors. Trincheras use
surfaces difficult to find and are generally not prepared. Most Hohokam house types have a post
hole pattern along the perimeter to support side walls and an internal roof support pattern that
generally varied by house shape. Trincheras house perimeters are ephemeral and do not have
post hole patterns. Hearths are found in most Hohokam house types and are generally well-fired.
The team sometimes found hearths in Trincheras residential structures. Entrances into Hohokam
dwellings are usually bulbous or parallel sided. Entries to Trincheras houses are not by
passageways.
Overall, Hohokam houses exhibit a much more substantial and permanent character than
those at Cerro de Trincheras. Despite the well-documented differences between the Hohokam of
the Phoenix basin and its peripheries, this study found that the typology originally formulated at
Snaketown can be applied with some modification to peripheral Hohokam territories. Thus, we
can say that the Tucson basin and the Papaguería architecture are much more like the Hohokam
This study then considers architectural similarities to the San Pedro Cochise. I conducted
164
the comparison to the San Pedro Cochise because researchers have postulated common origins
for the Hohokam and Trincheras from this group. The comparison yielded basic similarities
between the jacales and pithouses at Trincheras and the oval to circular structures of the San
Pedro Cochise. Both structure types generally have informal hearths, little evidence for structural
support.
The study of features at Cerro de Trincheras seems to reveal a basic paradox between the
monumentality of the site as a whole, and the apparent informality and impermanence of features
on the site. The visual impact of Cerro de Trincheras has lasted more than 500 years. The
inhabitants of Cerro de Trincheras invested more effort building a message of monumental scale
visible from surrounding areas than in the local constructions used in day to day life.
165